Retrieval-Augmented Generation (RAG) systems retrieve relevant documents from a corpus database to mitigate issues like hallucination, outdated knowledge, and limited domain coverage. While enhancing large language models (LLMs) performance, RAG also introduces a new attack surface: adversaries can inject trigger-embedded malicious documents into the corpus database, potentially causing the LLM to produce attacker-controlled outputs.

In post-disaster recovery scenarios, Unmanned Aerial Vehicles (UAVs) are widely deployed as aerial base stations to ensure deterministic communication services for time-sensitive (TS) flows requiring strict latency and reliability requirements. However, dynamic network topologies may render existing schedules invalid, and newly arriving flows can cause time-slot conflicts with ongoing transmissions, resulting in low resource utilization. To address these challenges, we propose a Depth-First Search-based Proximal Policy Optimization (DFS-PPO) approach for robust data flow routing and scheduling. The proposed method aims to balance link loads while satisfying reliability and latency constraints. A pruned DFS algorithm is used to determine the candidate routing set. To capture the spatial dependencies of all link states in the network, a Graph Attention Network (GAT) is employed to extract the global network state. In addition, a Transformer is adopted to model the nonlinear relationship between candidate routes and the network state. A PPO agent then makes routing decisions based on the high-level representation generated by the Transformer, enabling the model to adapt dynamically to changing network conditions and effectively mitigate bottlenecks. Simulation results demonstrate that the proposed DFS-PPO approach significantly outperforms benchmark algorithms in terms of network load balancing, reliability, and the number of successfully scheduled TS flows.

Time-sensitive missions in LEO satellite networks, such as fast position reporting, emergency communication, and surveillance, impose strict requirements on end-to-end delay and jitter. Early approaches like Satellite Formation Flying (SFF) relied on stable physical formations but suffered from tight coupling between orbits and services, limiting scalability and flexibility. This paper introduces a Virtual Satellite Formation (VSF) paradigm that decouples physical resources from service logic and dynamically selects virtual satellites to construct temporary formations. A two-layer network model is developed to quantify inter-satellite link stability, and an end-to-end deterministic transmission model is established by incorporating the characteristics of time-sensitive missions. Based on these models, we define a Deterministic Service Index (DSI) and formulate the corresponding optimization problem. To solve it, we propose a dynamic multi-objective evolutionary algorithm (DMO-VSF) that adapts to network dynamics in real time. Simulations in STK demonstrate that DMO-VSF reduces end-to-end delay by 15–18% and jitter by 35–40% compared with traditional routing, enabling resilient and deterministic services in large-scale constellations.

With the deepening of the Internet of Things (IOT) and industrial digital transformation, the core of positioning services is shifting from “serving people” to “connecting everything”, which poses a comprehensive challenge to indoor positioning technology in terms of high accuracy, low latency, low power consumption, and low cost. Traditional radio frequency positioning technology has shown many limitations in this context, while visible light positioning (VLP) technology has become a highly promising supplementary solution due to its unique advantages, such as the absence of authorized spectrum, no electromagnetic interference, high security, and the ability to balance lighting. However, traditional VLP methods heavily rely on accurate channel models and are difficult to cope with complex non-line-of-sight environments, resulting in increasingly prominent performance bottlenecks. In recent years, the rapid development of machine learning technology has provided a new paradigm for solving the above-mentioned problems. From the perspective of the IOT and edge computing, this paper systematically summarizes the latest progress of how machine learning can improve the performance of indoor VLP.We first elaborate on the architecture and basic principles of edge-oriented VLP systems. Then, a comprehensive review and comparison are performed on VLP methods based on traditional machine learning and deep learning. Moving on, we provide the analysis on how they improve system accuracy and robustness through data-driven approaches. Moreover, this article delves into the application and value of different learning paradigms, such as centralized learning, online learning, and federated learning in VLP systems. In addition, we have developed a multi-dimensional evaluation system that includes core positioning accuracy and edge performance indicators to comprehensively measure the feasibility of the system in practical deployment. Finally, we present the identified challenges and future research directions in this under-explored field from aspects of standardized scenario modeling, high generalization base models, dynamic environment robustness, heterogeneous terminal adaptation, and edge lightweight models.

Machine unlearning enables data holders to remove the contribution of their specified samples from trained models to protect their privacy. However, it is paradoxical that most unlearning methods require the unlearning requesters to first upload their data to the server as a prerequisite for unlearning. These methods are infeasible in many privacy-preserving scenarios where servers are prohibited from accessing users' data, such as federated learning (FL). In this paper, we explore how to implement unlearning under the condition of not uncovering the erasing data to the server. We propose Blind Unlearning (BlindU), which carries out unlearning using compressed representations instead of original inputs. BlindU only involves the server and the unlearning user: the user locally generates privacy-preserving representations, and the server performs unlearning solely on these representations and their labels. For the FL model training, we employ the information bottleneck (IB) mechanism. The encoder of the IB-based FL model learns representations that distort maximum task-irrelevant information from inputs, allowing FL users to generate compressed representations locally. For effective unlearning using compressed representation, BlindU integrates two dedicated unlearning modules tailored explicitly for IB-based models and uses a multiple gradient descent algorithm to balance forgetting and utility retaining. While IB compression already provides protection for task-irrelevant information of inputs, to further enhance the privacy protection, we introduce a noise-free differential privacy (DP) masking method to deal with the raw erasing data before compressing. Theoretical analysis and extensive experimental results illustrate the superiority of BlindU in privacy protection and unlearning effectiveness compared with the best existing privacy-preserving unlearning benchmarks.

With the accelerated development of large models and distributed training, the explosive growth of data volume has brought huge challenges to traditional data processing and machine learning algorithms. The inconsistency and accuracy of the data processing process will directly affect the analysis and decision-making effect of the data. In this paper, we investigate a data fusion framework based on a trustworthiness metric model, aiming to incorporate a credibility evaluation mechanism of data sources into the data fusion process. Firstly, we propose a trust measurement model based on dynamic Bayesian networks that is correlated with time factors, taking into account the impact of network interaction behavior on trust measurement. Secondly, we design a data security aggregation mechanism based on the trustworthy measurement model, which combines credibility measurement with the minimum spanning tree (MST) protocol to improve the network's perception performance. Finally, we conduct simulation experiments and real experimental bed tests separately, and the results showed that the proposed model has continuous trustworthiness measurement ability in dynamic uncertain network environments. On a large benchmark dataset, the proposed algorithm performs better than state-of-the-art methods in statistics, with a significant reduction rate of 36.6% in the computational cost of the control center, and a reduction rate of 23.8% and 34.7% in the communication and storage costs of the system, respectively.

The issue of association privacy leakage has become increasingly critical during data release and usage. However, traditional privacy protection techniques often struggle to address privacy leakage resulting from implicit associations within the data. In this paper, we propose a novel mechanism based on Cobweb Privacy to safeguard association privacy more comprehensively. Firstly, we design the concept of ϵ-Cobweb Privacy (ϵ-CP) specifically to address association privacy leakage. This concept extends the traditional notion of differential privacy by incorporating associated prior knowledge, thereby offering more effective and comprehensive protection of association privacy. We further demonstrate its privacy guarantees through the theoretical analysis of the relationship between ϵ-CP, differential privacy, and pufferfish privacy. Secondly, we quantify the privacy leakage problem mathematically and examine the utility privacy trade off under various priors. Additionally, we present a universal framework for association privacy protection in data aggregation scenarios using the ϵ-CP mechanism. Finally, this framework is integrated with three different noise addition methods and compared against mechanisms based on differential privacy and pufferfish privacy, and its utility is validated through experiments on both non-temporal and temporal real-world datasets. The results show that ϵ-CP provides distinct advantages in the utility privacy trade-off.

Many machine unlearning methods have been proposed recently to uphold users' right to be forgotten. However, offering users verification of their data removal post-unlearning is an important yet under-explored problem. Current verifications typically rely on backdooring, i.e., adding backdoored samples to influence model performance. Nevertheless, the backdoor methods can merely establish a connection between backdoored samples and models but fail to connect the backdoor with genuine samples. Thus, the backdoor removal can only confirm the unlearning of backdoored samples, not users' genuine samples, as genuine samples are independent of backdoored ones. In this paper, we propose a Self-supervised Model Seeding (SMS) scheme to provide unlearning verification for genuine samples. Unlike backdooring, SMS links user-specific seeds (such as users' unique indices), original samples, and models, thereby facilitating the verification of unlearning genuine samples. However, implementing SMS for unlearning verification presents two significant challenges. First, embedding the seeds into the service model while keeping them secret from the server requires a sophisticated approach. We address this by employing a self-supervised model seeding task, which learns the entire sample, including the seeds, into the model's latent space. Second, maintaining the utility of the original service model while ensuring the seeding effect requires a delicate balance. We design a joint-training structure that optimizes both the self-supervised model seeding task and the primary service task simultaneously on the model, thereby maintaining model utility while achieving effective model seeding. The effectiveness of the proposed SMS scheme is evaluated through extensive experiments on three representative datasets, utilizing various model architectures and exact and approximate unlearning benchmarks. The results demonstrate that SMS provides effective verification for genuine sample unlearning, effectively addressing the limitations of existing solutions.

Machine learning (ML) has gained significant adoption in Android malware detection to address the escalating threats posed by the rapid proliferation of malware attacks. However, recent studies have revealed the inherent vulnerabilities of ML-based detection systems to evasion attacks. While efforts have been made to address this critical issue, many of the existing defensive methods encounter challenges such as lower effectiveness or reduced generalization capabilities. In this paper, we introduce MalPurifier, a novel adversarial purification framework specifically engineered for Android malware detection. Specifically, MalPurifier integrates three key innovations: a diversified adversarial perturbation mechanism for robustness and generalizability, a protective noise injection strategy for benign data integrity, and a Denoising AutoEncoder (DAE) with a dual-objective loss for accurate purification and classification. Extensive experiments on two large-scale datasets demonstrate that MalPurifier significantly outperforms state-of-the-art defenses. It robustly defends against a comprehensive set of 37 perturbation-based evasion attacks, consistently achieving robust accuracies above 90.91%. As a lightweight, model-agnostic, and plug-and-play module, MalPurifier offers a practical and effective solution to bolster the security of ML-based Android malware detectors.

The development of Large Language Models (LLMs) faces a significant challenge: the exhaustion of publicly available fresh data. This is because training an LLM requires a large demand for new data. Federated learning emerges as a promising solution, enabling collaborative model to contribute their private data to LLM global model. However, integrating federated learning with LLMs introduces new challenges, including the lack of transparency and the need for effective unlearning mechanisms. Transparency is essential to ensuring trust and fairness among participants, while accountability is crucial for deterring malicious behaviour and enabling corrective actions when necessary. To address these challenges, we propose a novel blockchain-based federated learning framework for LLMs that enhances transparency, accountability, and unlearning capabilities. Our framework leverages blockchain technology to create a tamper- proof record of each model's contributions and introduces an innovative unlearning function that seamlessly integrates with the federated learning mechanism. We investigate the impact of Low-Rank Adaptation (LoRA) hyperparameters on unlearning performance and integrate Hyperledger Fabric to ensure the security, transparency, and verifiability of the unlearning process. Through comprehensive experiments and analysis, we showcase the effectiveness of our proposed framework in achieving highly effective unlearning in LLMs trained using federated learning. Our findings highlight the feasibility of integrating blockchain technology into federated learning frameworks for LLM.

Decentralized storage auditing approaches are designed to ensure data security in dishonest decentralized storage providers. However, the need for data updates introduces new challenges to the design of decentralized storage auditing approaches. Existing approaches can support dynamic auditing for updated files. Unfortunately, they can only deal with block-level updating, which is counter-intuitive and requires conversion from semantic changes to binary changes. Furthermore, existing dynamic auditing approaches require the recalculation of auxiliary auditing information (e.g., auditing authenticators) in data owners, which imposes unnecessary additional burdens on data owners, particularly those with constrained resources in decentralized storage environments. In this paper, we focus on image files and propose iAudit, an efficient pixel-level dynamic image auditing approach in decentralized storage. We first design a novel image authenticator with image pixels for efficient dynamic auditing, which combines convolution operations and polynomial commitment in authenticator construction. Additionally, we build an owner-free dynamic mechanism in dynamic decentralized storage auditing approach by utilizing zero-knowledge proof techniques. In this way, the dynamic operation overheads incurred by auditing can be completely eliminated from the data owners. A prototype of iAudit is implemented, and extensive experimental results demonstrate that iAudit outperforms state-of-the-art works, achieving over a 210× speedup for data owner in dynamic update phase.

Mobile edge caching plays a crucial role in traffic offloading for access networks. By storing frequently requested content items close to subscribers, it significantly reduces data retrieval latency, mitigates backhaul congestion, and alleviates the load on remote servers. Among various caching strategies, the two-tier Least Recently Used (2-LRU) policy has been widely adopted due to its efficient popularity-aware filtering capability while maintaining$\mathcal {O}(1)$computational complexity. However, conventional 2-LRU often allocates an equal number of entries to both LRU tiers, overlooking the potential cache hit ratio gains achievable through asymmetric tier-size configurations. Therefore, in this paper, we present a comprehensive analysis of 2-LRU under asymmetric tier sizing, and leverage the obtained insights to further guide performance optimization. In particular, we first construct a discrete-time Markov chain model to characterize the state transitions of 2-LRU and derive a closed-form expression for its cache hit probability, i.e., the hit ratio of its second-tier LRU ($C_{2}$). We then perform extensive simulations to validate accuracy of the proposed model and investigate the optimal size settings for the first-tier LRU ($C_{1}$). Building on the key implications of the associated results, we further propose 2LRU-$\Delta C_{1}$, an enhanced 2-LRU scheme that dynamically adjusts the size of$C_{1}$to accelerate the population of$C_{2}$with popular content data, thereby improving the cache hit ratio of$C_{2}$. Finally, we implement 2LRU-$\Delta C_{1}$in NS-3 and evaluate its performance against several baseline strategies, including LRU, 2-LRU ($C_{1}=C_{2}$), LFU, and a DRL-based policy. Corresponding results have confirmed the efficiency of our proposed scheme.

Traditional machine learning-based approaches to Internet of Things (IoT) traffic classification are hindered by the need for a large number of labeled flows in real-world network environments. To cope with this problem, some few-shot traffic classification approaches have been proposed to achieve accurate classification by comparing the similarity of flows to the given labeled flows. These approaches train neural networks to concurrently learn feature extraction and feature comparison meta-knowledge from sufficient labeled flows. However, they conflate these two types of meta-knowledge and ignore their differences. We propose MUFFIN, a meta-knowledge decoupling-based approach to few-shot IoT traffic classification. MUFFIN is based on our key insight that feature extraction and feature comparison meta-knowledge are two different types of meta-knowledge with distinct learning goals. Therefore, MUFFIN decouples their learning in the meta-knowledge phase and introduces a masked autoencoder as its feature extractor to enhance the learning of feature extraction meta-knowledge. Extensive experiments conducted on three publicly available datasets demonstrate the classification ability and robustness of MUFFIN. Furthermore, a comparison with five state-of-the-art few-shot traffic classification approaches reveals that MUFFIN outperforms them in classification accuracy and requires only 20 labeled flows.

Accurate location information is an essential prerequisite for location-based service and operation safety in railways. Ultra wide band (UWB) fused with inertial measurement unit (IMU) has been thought as a promising localization technique for trains. However, due to trains’ high speed, ranging results from UWBs are seriously influenced by Doppler effect. Most existing work focused on Doppler effect mitigation in single round of communication, but ignored its negative impact on two-way time of flight (ToF) and fusion performance. In this paper, we propose an adaptive fusion model with ranging rectifications for high-speed trains. Firstly, two-way ToF ranging errors caused by Doppler effect are studied. Secondly, Doppler effect is mitigated effectively, with prior information of trains’ states in time series. Finally, an adaptive UWB/IMU fusion scheme is proposed, where residual errors of UWB observations after Doppler effect mitigation are modeled by a closed-form expression. Simulation results show that, regarding single time-slot localization, the proposed ranging rectification algorithm outperforms four typical algorithms, with gains of 78.42%, 60.22%, 58.09%, and 28.38%, in terms of RMSE. Regarding continuous localization with temporal correlation, the proposed adaptive fusion model achieves gains of 75.07%, 73.27%, 53.52% and 70.02%. Furthermore, its effectiveness is verified in practical platform.

The joint design of sensing, communication, computing, and control (SC3) is crucial for supporting environment-aware Industrial Internet of Things (IIoT) applications. Considering the uncontrollable wireless propagation environments and limited spectrum resources, wireless communication performance often becomes the primary design bottleneck for such an integrated system. To address this challenge, this paper presents a design framework for reconfigurable intelligent surface (RIS)-enhanced semantic-aware SC3networks, where RIS and semantic communication technologies are employed to improve wireless communication efficiency. To facilitate real-time closed-loop control, we further formulate a weighted sum execution latency minimization problem, while imposing constraints on maximum execution latency and energy consumption of individual IoT device, as well as minimum information entropy to meet specific control requirements measured by linear quadratic regulator cost. In addition, the design framework aims at optimizing bandwidth allocation, RIS phase shift matrix, time scheduling, transmit power, and CPU-cycle frequency for IoT devices and the base station (BS). To handle the coupled multi-dimensional optimization variables, the block coordinate descent method is utilized to decompose the formulated problem into more tractable subproblems, which are then solved using a penalty-function-based approach and geometric programming technique. Simulation results demonstrate the performance advantages achieved by our proposed method compared to several benchmark approaches. Additionally, we explore the impact of various parameters on SC3systems, offering deeper insights and meaningful research observations.

Semantic communication has emerged as a pivotal technology for future wireless communication, where privacy preservation plays a critical role. However, existing privacy-preserving semantic communication methods have a negative impact on communication performance. In this paper, we focus on fine-grained privacy-preserving semantic communication to achieve a better performance-privacy trade-off. Specifically, we formalize it as an optimization problem and propose a Dual Information Bottleneck (Dual-IB) method to solve it. Firstly, Dual-IB extracts task-relevant information from input and compresses task-irrelevant information. Secondly, it minimizes sensitive attributes within the task-relevant representation by information bottleneck. This mitigates the risks of privacy exposure while maintaining the task performance. We validate the effectiveness of our method using an eavesdropping attack that attempts to extract private information in the transmission. The experimental results demonstrate that our method not only achieves a task performance of approximately 97%, but also reduces eavesdropping accuracy by 68%, effectively addressing a critical limitation of existing approaches.

Uncovering cause-and-effect mechanisms from data is fundamental to scientific progress. While large language models (LLMs) show promise for enhancing causal discovery (CD) from unstructured data, their application to the increasingly prevalent multimodal setting remains a critical challenge. Even with the advent of multimodal LLMs (MLLMs), their efficacy in multimodal CD is hindered by two primary limitations: (1) difficulty in exploring intra- and inter-modal interactions for comprehensive causal variable identification; and (2) insufficiency to handle structural ambiguities with purely observational data. To address these challenges, we propose MLLM-CD, a novel framework for multimodal causal discovery from unstructured data. It consists of three key components: (1) a novel contrastive factor discovery module to identify genuine multimodal factors based on the interactions explored from contrastive sample pairs; (2) a statistical causal structure discovery module to infer causal relationships among discovered factors; and (3) an iterative multimodal counterfactual reasoning module to refine the discovery outcomes iteratively by incorporating the world knowledge and reasoning capabilities of MLLMs. Extensive experiments on both synthetic and real-world datasets demonstrate the effectiveness of the proposed MLLM-CD in revealing genuine factors and causal relationships among them from multimodal unstructured data. The implementation code and data are available at https://github.com/JinLi-i/MLLM-CD.

With the rapid development of Artificial Intelligence, fake data are being generated and circulated on an unprecedented scale. While such data may be entertaining in some contexts, data collectors are generally reluctant to accept them due to concerns about service quality and security risks. A central challenge, therefore, is how to incentivise contributors to provide authentic data. Existing research has mainly focused on regulating the use of fake data, with limited attention to designing mechanisms that motivate contributors to share real data. In this paper, we propose a pricing strategy with an embedded incentive mechanism that ensures contributors obtain higher financial benefits from sharing real data rather than fake data. We formulate the problem as a Stackelberg game, where the data collector acts as the leader and contributors are the followers. To enhance reliability, we integrate a peer-review method to help verify data authenticity, which in turn informs the incentive-compatible pricing design. We establish the existence and uniqueness of the equilibrium solution, and further conduct numerical simulations to demonstrate the equilibrium-seeking process and the influence of key parameters.

With the extensive deployment and application of Internet of Things (IoT) devices, vulnerable edge nodes have emerged as primary targets for Advanced Persistent Threat (APT) attacks. Attackers compromise IoT terminal devices to establish an initial foothold and subsequently exploit lateral movement techniques to progressively infiltrate core business networks. Prior investigation methods struggle with fragmented threat intelligence and sparse attack samples in heterogeneous audit logs, resulting in incomplete attack chain reconstruction and high false positives. We propose a novel approach to APT attack investigation, FALCON, which captures complex causal relationships between entities from discrete audit logs and constructs cross-domain provenance graphs, enabling rapid and accurate identification of potential APT activities. FALCON trains an adaptive edge-side local model with cross-domain behavior sequences containing extensive and remote contextual information, and employs a bidirectional transformer pre-trained model to learn latent representations from unlabeled sequences. To the best of our knowledge, FALCON is the first APT investigation method to conduct causal provenance based on cross-domain audit logs while ensuring privacy protection. The experimental results demonstrate that FALCON effectively detects APT attacks with accuracy 99.71% and reconstructs attack scenarios with accuracy 87.4%.

The sixth generation (6G) communication posed higher requirements for the communication system regarding accurate semantic transmission. The existing studies about semantic communication principally concentrate on tackling task-oriented problems, rather than directly design modality-oriented system which is more generic and adaptive to different tasks. To improve the flexibility and robustness of communication system, we propose a Multimodality-Oriented Unified Semantic Communication SystEm (MOUSSE) based on contrastive learning. MOUSSE is designed to firstly orient modality then matches different modalities combination up to various tasks. Existing task-oriented philosophy primarily considers tasks whilst restricting modal versatility. MOUSSE could also intake and output various tasks with multiple modalities while transmit them in a concise and unified representation. Specifically, the system consists of structure-symmetric twin encoder-decoder for modality unification, which cascades joint source-channel coding (JSCC) module and contrastive learning based alignment module. Finally, the experiments verify the validity of proposed MOUSSE by quantitative results from different modalities with their respective tasks. The reliability and robustness are also improved from the point of entire communication system view.

Incomplete scenario is a prevalent, practical, yet challenging setting in Multimodal Recommendations (MMRec), where some item modalities are missing due to various factors. Recently, a few efforts have sought to improve the recommendation accuracy by exploring generic structures from incomplete data. However, two significant gaps persist: 1) the difficulty in accurately generating missing data due to the limited ability to capture modality distributions; and 2) the critical but overlooked visibility bias, where items with missing modalities are more likely to be disregarded due to the prioritization of items' multimodal data over user preference alignment. This bias raises serious concerns about the fair treatment of items. To bridge these two gaps, we propose a novel Modality-Diffused Counterfactual (MoDiCF) framework for incomplete multimodal recommendations. MoDiCF features two key modules: a novel modality-diffused data completion module and a new counterfactual multimodal recommendation module. The former, equipped with a particularly designed multimodal generative framework, accurately generates and iteratively refines missing data from learned modality-specific distribution spaces. The latter, grounded in the causal perspective, effectively mitigates the negative causal effects of visibility bias and thus assures fairness in recommendations. Both modules work collaboratively to address the two aforementioned significant gaps for generating more accurate and fair results. Extensive experiments on three real-world datasets demonstrate the superior performance of MoDiCF in terms of both recommendation accuracy and fairness. The code and processed datasets are released at https://github.com/JinLi-i/MoDiCF.

With the increasing prevalence of Web-based platforms handling vast amounts of user data, machine unlearning has emerged as a crucial mechanism to uphold users' right to be forgotten, enabling individuals to request the removal of their specified data from trained models. However, the auditing of machine unlearning processes remains significantly underexplored. Although some existing methods offer unlearning auditing by leveraging backdoors, these backdoor-based approaches are inefficient and impractical, as they necessitate involvement in the initial model training process to embed the backdoors. In this paper, we propose a TAilored Posterior diffErence (TAPE) method to provide unlearning auditing independently of original model training. We observe that the process of machine unlearning inherently introduces changes in the model, which contains information related to the erased data. TAPE leverages unlearning model differences to assess how much information has been removed through the unlearning operation. Firstly, TAPE mimics the unlearned posterior differences by quickly building unlearned shadow models based on first-order influence estimation. Secondly, we train a Reconstructor model to extract and evaluate the private information of the unlearned posterior differences to audit unlearning. Existing privacy reconstructing methods based on posterior differences are only feasible for model updates of a single sample. To enable the reconstruction effective for multi-sample unlearning requests, we propose two strategies, unlearned data perturbation and unlearned influence-based division, to augment the posterior difference. Extensive experimental results indicate the significant superiority of TAPE over the state-of-the-art unlearning verification methods, at least 4.5x efficiency speedup and supporting the auditing for broader unlearning scenarios.

Despite the rapid growth of electric vehicles (EVs), charging remains a time-consuming issue that requires effective management. An important solution uses digital twin (DT) technology, which acts as a virtual agent for EVs in the digital space. DT can analyze real-time vehicle data to develop optimal charging schedules and reserve charging providers in advance through the vehicular network, leading to more efficient charging processes. However, the vehicular network exposes the automated reservation process of the DT to security attacks. Additionally, there is a risk that the actual charging process may deviate from the scheduled requirements set by the DT, resulting in wasted charging resources. To address these issues, this article proposes a secure and efficient charging reservation scheme based on DT technology. To prevent malicious attacks, we first design a secure and privacy-preserving reservation authentication protocol using the extended Chebyshev chaotic maps, taking into account the computational resources of the EV. Furthermore, we develop a reputation mechanism to evaluate and incentivize the charging behavior of EVs. Formal verification and further discussions are conducted to show diverse security functionalities of the proposed scheme can be achieved. We evaluate that the proposed scheme outperforms existing schemes in terms of computation and communication overheads, while also assessing the impact of EV charging behavior on reputation and charging level.

Federated learning (FL) allows multiple devices to train a high-performance global model cooperatively without sharing their private data. One of the key challenges in FL for Internet of Things (IoT) scenarios is that the statistical heterogeneity in local data distribution for IoT devices, which will cause the low quality of local models, degrading the performance of the global model. To address this problem, focusing on improving the quality of local models, we propose a match-based hierarchical FL framework (FedAvg-Match), in which two IoT devices with complementary datasets are formed as a training group to reduce the influence of data heterogeneity. By introducing Earth mover’s distance for data distribution, we design an improved Gale-Shapley algorithm with time complexity$O(N^{2}\log N)$for IoT device grouping, which can obtain a stable matching. Experimental results show that, compared to the ungrouped FedAvg algorithm, the proposed FedAvg-Match method significantly improves both the accuracy of the global model and the convergence speed of the training process, while also reducing communication costs.

The integration of edge computing with deep neural networks (DNNs) is crucial for intelligent industrial cyber-physical systems. Typically, deploying DNNs on heterogeneous edge devices relies on methods like model compression and partitioning. However, these approaches often result in homogeneous models across devices. This homogeneity limits the collective capability of edge computing systems, particularly in terms of generalization to diverse data distributions and adaptation to dynamic industrial environments. In this work, we propose to treat each DNN on an edge device as an independent model, aggregating their capabilities via ensemble learning to enhance generalization and dynamic adaptability. To realize this, we introduce the Adaptive Ensemble Knowledge Distillation Framework (AEKDF), combining cloud-based model training with edge computing based collaborative inference. In the cloud, AEKDF develops an enhanced Born Again Network that generates diverse, lightweight models tailored to specific edge devices through knowledge distillation. This process ensures model diversity which is critical to effective ensemble learning. On the edge, AEKDF employs an adaptive ensemble technique that aggregates prediction logits across devices, enabling rapid adaptation to changing environments and maintaining inference efficiency. Our extensive evaluations conducted on a realistic prototype demonstrate the substantial boost in predictive performance achieved by our AEKDF, showcasing a 4% to 10% accuracy improvement on the CIFAR-100 compared to conventional single-model approaches, while maintaining low latency.

Depending on large-scale devices, the Internet of Things (IoT) provides massive data support for resource sharing and intelligent decision, but privacy risks also increase. As a popular distributed learning framework, Federated Learning (FL) is widely used because it does not need to share raw data while only parameters to collaboratively train models. However, Federated Learning is not spared by some emerging attacks, e.g., membership inference attack. Therefore, for IoT devices with limited resources, it is challenging to design a defense scheme against the membership inference attack ensuring high model utility, strong membership privacy and acceptable time efficiency. In this paper, we propose MemDefense, a lightweight defense mechanism to prevent membership inference attack from local models and global models in IoT-based FL, while maintaining high model utility. MemDefense adds crafted pruning perturbations to local models at each round of FL by deploying two key components, i.e., parameter filter and noise generator. Specifically, the parameter filter selects the apposite model parameters which have little impact on the model test accuracy and contribute more to member inference attacks. Then, the noise generator is used to find the pruning noise that can reduce the attack accuracy while keeping high model accuracy, protecting each participant's membership privacy. We comprehensively evaluate MemDefense with different deep learning models and multiple benchmark datasets. The experimental results show that lowcost MemDefense drastically reduces the attack accuracy within limited drop of classification accuracy, meeting the requirements for model utility, membership privacy and time efficiency.

Multi-path transmission enables load balancing and improves network performance in data center networks (DCNs). It increases the possibility of network congestion and makes traditional network traffic engineering methods inefficient due to the uneven distribution of network traffic in data centers. In this paper, we present a reliable and efficient Disjoint paths based Multi-Path Transmission scheme (DMPT) that selects distributed requests through topology awareness. Firstly, we propose disjoint path construction algorithms through rigorous theoretical proof, aiming at the different transmission requirements of DCNs. Secondly, we offer an optimal solution to the disjoint multi-path selection problem, which is aimed at the trade-off between link load and transmission time. Furthermore,DMPTcan split the flow over multiple transmission paths based on the link status. Finally, extensive experiments are executed forDMPTon a novel EHDC of DCN that is based on exchanged hypercube. The experimental results show thatDMPTcan reduce the average running time by 18.6%, and the average path length is close to the optimal path. Furthermore, it achieves significant improvements in balancing network link traffic and facilitating deployment, which also reflects the advantages of topology aware multiplexing in practice.

Self-Supervised Learning (SSL) excels in utilizing unlabeled data for feature representation learning. However, recent studies have revealed that SSL is vulnerable to data poisoning-based backdoor attacks. To remove backdoored samples from the SSL training dataset, model optimization methods often fine-tune a trained model by contrasting the training dataset with a reserved clean dataset. This contrastive training effectively marginalizes backdoored samples from the distribution of benign ones,if and only ifboth the reserved clean dataset and the training dataset are from the same data distribution. However, presuming identical distributions between the web-scraped data and reserved data is impractical. To address this impractical assumption, our proposed Bootstrapped Dual SetPurification (AUTO) method distinguishes backdoored from benign samples by contrasting a mined ‘positive set’ and a mined ‘negative set’ within the training dataset itself. We exploit the resistance of backdoored samples in data mixing to mine a highly poisoned ‘positive set’ and a minimally poisoned ‘negative set’. Besides, AUTO mitigates unstable detection performance within different optimization steps by continuously refining the dual sets by the optimized model, enhancing the model's poison distinguishability from consistently improving supervision signals. Our extensive experiments on Cifar10, Cifar100, and Imagenet100 against existing data poisoning SSL backdoor attacks demonstrate AUTO's superiority in detection performance over all existing defenses.

Recently, self-supervised learning has garnered significant attention for its ability to extract high-quality features from unlabeled data. However, existing research indicates that backdoor attacks can pose significant threats to self-supervised learning. Additionally, due to the substantial training overhead, traditional backdoor detection methods in supervised learning, such as meta-learning, are not well-suited for self-supervised learning. Current backdoor detection methods for self-supervised learning can only defend against backdoor attacks triggered by patch. To address this challenge, we proposes DBSSL, a scheme that can efficiently detect backdoor attacks in self-supervised learning models. In DBSSL, we connect backdoor attacks with adversarial attacks and use adversarial perturbations for detection. Specifically, we first conduct targeted adversarial attacks on the samples, classifying them into different classes. Then, we employ the Z-Score method to analyze these adversarial perturbations. If outliers are present, we can infer that the self-supervised learning model has been backdoored. Unlike previous works, our scheme is better suited for downstream users with constrained computation resources and exhibits excellent detection capabilities for backdoor attacks caused by patch-type or global triggers. Abundant theoretical analysis has demonstrated the feasibility of our scheme. Moreover, extensive experiments show that our method performs well in detecting prevalent backdoor attacks in self-supervised learning, achieving detection accuracy exceeding 95%.

Deep neural networks (DNNs) are known to be susceptible to various malicious attacks, such as adversarial and backdoor attacks. However, most of these attacks utilize additive adversarial perturbations (or backdoor triggers) within an$L_{p}$-norm constraint. They can be easily defeated by image preprocessing strategies, such as image compression and image super-resolution. To address this limitation, instead of using additive adversarial perturbations (or backdoor triggers) in the pixel space, this work revisits the design of adversarial perturbations (or backdoor triggers) from the perspective of color space and conducts a comprehensive analysis. Specifically, we propose a color space backdoor attack and a color space adversarial attack where the color space shift is used as the trigger and perturbation. To find the optimal trigger or perturbation in the black-box scenario, we perform an iterative optimization process with the Particle Swarm Optimization algorithm. Experimental results confirm the robustness of the proposed color space attacks against image preprocessing defenses as well as other mainstream defense methods. In addition, we also design adaptive defense strategies and evaluate their effectiveness against color space attacks. Our work emphasizes the importance of the color space when developing malicious attacks against DNN and urges more research in this area.

Defending Edge intelligence-based Internet of Things (EIoT) systems by controlling malware propagation has become a critical issue. Herein, we meet new challenges brought by multiple attackers and defenders for effective mitigation of malware propagation in EIoT. To explore the dynamic changes of malware propagation, a state transition diagram of IoT nodes is proposed to describe the mutual changes of five states: infected, active, dormant, isolated, and hardened, and differential equations for each state are established. We then build a mean-field game model representing the interactions between multiattackers and multidefenders in the EIoT malware propagation defense environment. We further convert the problem of solving the game into an MDP (Markov Decision Process) and propose a distributed algorithm called MFGA3C (Mean-Field Game-based Asynchronous Advantage Actor-Critic) that combines mean-field game and multiagent asynchronous advantage actor-critic to learn the optimal defense policy. Finally, we compare our algorithm MFGA3C with other benchmark algorithms in the EIoT malware propagation countermeasure environment. Experimental results show that our MFGA3C algorithm dominates the average reward, total reward, and the number of successful defenses under three typical EIoT-environmental conditions, which indicates that MFGA3C faster and more robustly learns the optimal malware propagation defense policy, contributing to protecting EIoT systems.

Machine unlearning allows data owners to erase the impact of their specified data from trained models. Unfortunately, recent studies have shown that adversaries can recover the erased data, posing serious threats to user privacy. An effective unlearning method removes the information of the specified data from the trained model, resulting in different outputs for the same input before and after unlearning. Adversaries can exploit these output differences to conduct privacy leakage attacks, such as reconstruction and membership inference attacks. However, directly applying traditional defenses to unlearning leads to significant model utility degradation. In this article, we introduce a Compressive Representation Forgetting Unlearning scheme (CRFU), designed to safeguard against privacy leakage on unlearning. CRFU achieves data erasure by minimizing the mutual information between the trained compressive representation (learned through information bottleneck theory) and the erased data, thereby maximizing the distortion of data. This ensures that the model's output contains less information that adversaries can exploit. Furthermore, we introduce a remembering constraint and an unlearning rate to balance the forgetting of erased data with the preservation of previously learned knowledge, thereby reducing accuracy degradation. Theoretical analysis demonstrates that CRFU can effectively defend against privacy leakage attacks. Our experimental results show that CRFU significantly increases the reconstruction mean square error (MSE), achieving a defense effect improvement of approximately 200% against privacy reconstruction attacks with only 1.5% accuracy degradation on MNIST.

Machine unlearning has become a significant research topic on a global scale due to the increasing importance of privacy protection, particularly in light of the right to be forgotten legislation. Although many solutions are proposed, the current mainstream centralized machine unlearning studies are not feasible in federated learning (FL), where the server has no access to any users’ unlearning samples. In this paper, we aim to tackle thefederated unlearningproblem by proposing a Federated Unlearning (FedU) scheme via a user-side influence approximation forgetting method, thereby eliminating the need to share raw data with the server. In FedU, only users who have unlearning needs execute the influence approximation forgetting, while other users and the server just conduct the same operations as they did in FL. The proposed influence approximation forgetting method achieves unlearning by estimating the influence of the erased samples relying on only the user's local data and eliminating this influence from the model. However, the model utility is still negatively influenced by directly removing the influence estimation. To mitigate the side effects of unlearning, we propose a utility preservation method that simultaneously trains the unlearned model based on the unlearning requesters’ remaining local dataset. We design an adaptive optimization method to balance the forgetting and utility preservation effectiveness optimally during the unlearning process. Extensive evaluations on three representative public datasets demonstrate that our proposed method significantly outperforms state-of-the-art methods in both effectiveness and efficiency, avoiding more than 3% accuracy degradation when the number of unlearning requesters is large.

Federated Learning (FL) represents a distributed machine learning approach, enabling the joint training of a global model through the aggregation of gradients from participating clients without necessitating the exchange of raw data. Prior research has explored methods for verifying the correctness of aggregation in this context and mitigating the overhead associated with the verification process. Nonetheless, the advent of Large Language Models (LLMs), with their parameters numbering in the billions, presents ongoing challenges in devising efficient verification mechanisms in FL for large models. In this paper, we propose an innovative Efficient Verifiable Federated Learning scheme${\sf LaVFL}$, which focusing on addressing the verification challenges incurred by LLM. Specifically, we propose an efficient layer-by-layer verification approach for LLMs by designing a Convolution Gradient Compression (CGC) method without compromising model accuracy. Additionally, to minimize computational and communication overheads, we propose an efficient verification strategy PGS, namely, a Probabilistic Gradient Sampling strategy, which aims to reduce the gradient dimensions for each round of verification while ensuring a high probability of comprehensive verification. We implement a prototype of${\sf LaVFL}$, and extensive experimental results demonstrate that${\sf LaVFL}$achieves over a$300 \times$speedup in the aggregation verification phase and reduces communication overheads by more than 75%, compared to VeriFL under the same experimental setup.

Deep learning (DL) enabled semantic communications leverage DL to train encoders and decoders (codecs) to extract and recover semantic information. However, most semantic training datasets contain personal private information. Such concerns call for enormous requirements for specified data erasure from semantic codecs when previous users hope to move their data from the semantic system. Existing machine unlearning solutions remove data contribution from trained models, yet usually in supervised sole model scenarios. These methods are infeasible in semantic communications that often need to jointly train unsupervised encoders and decoders. In this paper, we investigate the unlearning problem in DL-enabled semantic communications and propose a semantic communication unlearning (SCU) scheme to tackle the problem. SCU includes two key components. Firstly, we customize the joint unlearning method for semantic codecs, including the encoder and decoder, by minimizing mutual information between the learned semantic representation and the erased samples. Secondly, to compensate for semantic model utility degradation caused by unlearning, we propose a contrastive compensation method, which considers the erased data as the negative samples and the remaining data as the positive samples to retrain the unlearned semantic models contrastively. Theoretical analysis and extensive experimental results on three representative datasets demonstrate the effectiveness and efficiency of our proposed methods.

Increasing attention is being paid to machine unlearning, which supports individuals’ “right to be forgotten.” While most studies focus on the efficiency and effectiveness of unlearning algorithms, the evaluation of machine unlearning effectiveness remains underexplored. Offering robust evaluation services for unlearning is critical, not only to uphold privacy legislation but also to assess and improve existing unlearning methods. Lots of existing methods employ backdoor methods to evaluate unlearning effectiveness, which can only verify the unlearning effect of backdoored samples and negatively impact the model utility as they need to embed backdoors into the model first. In this paper, we propose an evaluating machine unlearning (EMU) method, which aims to evaluate the effectiveness of unlearning and verify data removal without the aforementioned adverse effects. Machine unlearning inherently creates a difference on the model before and after unlearning. The model difference contains information about the unlearned samples, which can be extracted through reconstruction models for unlearning effectiveness evaluation. To efficiently generate the model differences as input for evaluation, we simulate the model changes based on the influence function theory. Additionally, we design a multi-task information bottleneck structure to enhance the scalability of EMU and simplify the analysis of different learning tasks. We provide a theoretical analysis of how the similarity between erased and remaining samples, as well as task types, affects the extent of unlearning—factors that have been largely overlooked. Extensive experiments on various model architectures and representative datasets confirm our analysis, demonstrating the effective evaluation for unlearning without any degradation in the service model utility.

Internet of Medical Things (IoMT) has garnered significant research attention from both academic and medical institutions. However, the sensitive medical data involved in IoMT raises security and privacy concerns. To mitigate these, ring signature has surfaced as a proper solution, which offers unforgeability and anonymity. Unfortunately, most multi-signer ring signature schemes require a predetermined number of signers and are difficult to adjust dynamically. Additionally, traditional ring signatures have limited adaptability for IoMT due to their reliance on a single entity. It is challenging to effectively define different signature permissions for users of various entities, such as patients and doctors. Therefore, research focusing on constructing a dynamic multi-signer ring signature for multi-party participation remains a critical and ongoing challenge. In this paper, we present ASMR, an anonymity-enhanced sequential multi-signer ring signature scheme for secure medical data sharing in IoMT. ASMR contains two different rings, PR and DR, for patients and doctors, respectively. It allows patients in PR to anonymously sign their electronic healthcare record (EHR) owned by doctors in DR, overcoming the reliance on a single entity in existing approaches while enhancing the anonymity of the signature. Meanwhile, ASMR introduces the concept of signature chaining, allowing multiple users in DR to co-sign information in sequence. In addition, it ensures that each generated signature is traceable, offering a transparent system. We also formally prove the security of ASMR in the random oracle model. Comprehensive performance evaluations indicate that ASMR excels in both computational and storage overhead. In the best case, computational overhead is reduced by approximately 4.7×-61.7×, while storage overhead is reduced by approximately 26.7×-212.8× compared to prior arts.

In unsupervised graph anomaly detection, existing methods usually focus on detecting outliers by learning local context information of nodes, while often ignoring the importance of global context. However, global context information can provide more comprehensive relationship information between nodes in the network. By considering the structure of the entire network, detection methods are able to identify potential dependencies and interaction patterns between nodes, which is crucial for anomaly detection. Therefore, we propose an innovative graph anomaly detection framework, termed CoCo (Context Correlation Discrepancy Analysis), which detects anomalies by meticulously evaluating variances in correlations. Specifically, CoCo leverages the strengths of Transformers in sequence processing to effectively capture both global and local contextual features of nodes by aggregating neighbor features at various hops. Subsequently, a correlation analysis module is employed to maximize the correlation between local and global contexts of each normal node. Unseen anomalies are ultimately detected by measuring the discrepancy in the correlation of nodes’ contextual features. Extensive experiments conducted on six datasets with synthetic outliers and five datasets with organic outliers have demonstrated the significant effectiveness of CoCo compared to existing methods.

Massive urban-scale vehicle trajectories benefit various downstream applications. However, trajectories collected from existing sensing systems are often incomplete, necessitating the recovery of coarse-grained trajectories. Considering that mobility knowledge learned from a single system is less representative of all vehicles or covers only partial road segments, it becomes essential to combine diverse data from multiple systems to support trajectory recovery. Therefore, we learn the impacts of mobility intentions and dynamic traffic conditions on the movement of vehicles from trajectories aggregated across different systems to recover their travel routes on unobservable road intersections. Nonetheless, aggregating raw data across multiple systems raises privacy concerns. This data isolation compounds challenges in acquiring comprehensive mobility intentions and traffic conditions, thereby impairing recovery performance. In this paper, we proposeCrossTrace, a two-stage framework for privacy-aware cross-system trajectory recovery: in theTraffic Condition Inferencestage, a Split Learning pipeline with a multi-view graph neural network is utilized to infer complete traffic conditions for all road segments; in theTrajectory Recoverystage, a Federated Learning pipeline with dedicated modules is utilized to recover missing points by fusing inferred traffic conditions and mobility intentions. Extensive experiments on two large-scale trajectory datasets demonstrate thatCrossTraceoutperforms all alternative schemes.

Reconfigurable Intelligent Surface (RIS)-assisted uncrewed Aerial Vehicle (UAV) communications have been realized as essential to space-air-group system integration in the 6 G technology landscape. Trajectory planning plays a crucial role in RIS-assisted UAV communications to face the challenges of UAV’s limited power capacities and dynamic wireless channels. Existing solutions assume complete channel state information, focus on single-rotor UAVs, and rely heavily on time-consuming training processes for machine learning; thus, they lack applicability to deal with highly dynamic real-world scenarios. To fill these research gaps, we aim to characterize RIS-assisted UAV communications and design responsive and accurate UAV trajectory planning algorithms in this paper. We first develop a communication model with incomplete information and an energy consumption model for quadrotor UAVs. We then formulate UAV trajectory planning as an optimization problem to minimize UAV’s energy consumption while maintaining communication throughput. To solve this problem, we design an acceleration framework,FedX, for reinforcement learning (RL) solvers and present two fast trajectory planning algorithms, FedSAC and FedPPO, as instantiations of theFedXframework. Our evaluation results indicate that the proposed framework is effective and efficient–more than 3 times faster with 5 agents and 7 times faster with 10 agents than standard RL algorithms, making it suitable for using RL solvers within wireless networks and mobile computing environments. We also discuss and identify the pros and cons of our proposed framework.

Malware in the Internet of Things (IoT) is prone to contaminating various IoT end-points through network communication and information transfer, leading to surreptitious privacy leakage and data theft. The existing privacy-preserving approaches including data masking, anonymization, and differential privacy always lack the consideration of strategic interactions among rational agents. Inspired by Bayesian games, we model incomplete stochastic games between IoT end-points and edge nodes in edge intelligence (EI)-enabled IoT systems to conduct probability analysis for predicting and defending privacy leakage caused by malware infection. It is notable that the posterior probability is defined based on the Bayes’ rule to reflect the statistical inference of incomplete privacy leakage information. Such a method can intrinsically characterize the actual situations of IoT end-points. Further, we propose a novel privacy preservation optimization approach named Bayesian advantage actor critic (BA2C) for the practical implementation of optimization decision in EI-enabled IoT privacy-preserving systems. Eventually, we conduct experimental simulations to understand the most effective parameters in decision-making among the successful detection rate, successful infection rate, and false alarm rate. We also compare traditional algorithms and validate the efficacy of the proposed approach.

In Internet of Things (IoT) networks, the IoT sensors collect valuable raw data required to sustain Artificial Intelligence (AI) based networks operation. AI models are data-driven as they use the data to make accurate network security, management, and operational decisions. Unfortunately, the sensors are deployed in harsh environments which affects the sensor behaviour and eventually the networks' operations. Further, IoT devices are typically vulnerable to a range of malicious events. Therefore, IoT sensor's correct operation including resilience to failure is essential for sustained operations. Naturally, the state variables of time-series data can be changed, i.e., the data streams generated in these situations can be incorrect, incomplete or missing, and sparse presenting a significant challenge for real-time decision-making ability of AI models to make explainable and intelligent management and control decisions. In this paper, we aim to alleviate this fundamental problem to predict the missing and faulty reading correctly so that the decision-making ability of the AI models should not deteriorate in the presence of incorrect, missing, and highly imbalanced data sets. We use a novel approach using fuzzy-based information decomposition to recover the missed data values. We use three data sets, and our preliminary results show that our approach effectively recovers the missed or compromised data samples and help AI models in making accurate decision. Finally, the limitations and future work of this research have been discussed.

This work investigates the problem of efficiently learning discriminative low-dimensional (LD) representations of multiclass image objects. We propose a generic end-to-end approach that jointly optimizes sparse dictionary and convolutions for learning LOW-dimensional discriminative image representations, named SparConvLow, taking advantage of convolutional neural networks (CNNs), dictionary learning, and orthogonal projections. The whole learning process can be summarized as follows. First, a CNN module is employed to extract high-dimensional (HD) preliminary convolutional features. Second, to avoid the high computational cost of direct sparse coding on HD CNN features, we learn sparse representation (SR) over a task-driven dictionary in the space with the feature being orthogonally projected. We then exploit the discriminative projection on SR. The whole learning process is consistently treated as an end-to-end joint optimization problem of trace quotient maximization. The cost function is well-defined on the product of the CNN parameters space, the Stiefel manifold, the Oblique manifold, and the Grassmann manifold. By using the explicit gradient delivery, the cost function is optimized via a geometrical stochastic gradient descent (SGD) algorithm along with the chain rule and the backpropagation. The experimental results show that the proposed method can achieve a highly competitive performance with the state-of-the-art (SOTA) image classification, object categorization, and face recognition methods, under both supervised and semi-supervised settings. The code is available at https://github.com/MVPR-Group/SparConvLow.

Blockchain technology, known for its immutable distributed ledger, which greatly improves the credibility of data management in various applications. However, the immutability may result in erroneous data being permanently stored in the blockchain, affecting the security of the blockchain system. To address this issue, redactable blockchain is proposed to allow flexibly modifications of erroneous data in blockchain. Despite their promise, current redactable blockchain solutions often fall short in balancing two critical aspects: decentralization and data accountability. In this paper, we propose a Decentralized, Accountable and Redactable Blockchain solution (DARB). Our solution aims to mitigate the risks associated with central authority while ensuring secure accountability in data modification processes. Decentralization is achieved by using authorized authorities instead of the central authority, utilizing decentralized ciphertext policy attribute-based encryption. Additionally, it incorporates traceable ring signatures, which enable the authorized authorities to collaborate effectively in holding edited data accountable and tracing the identity of malicious modifiers. The security analysis and experimental results demonstrate that the DARB scheme successfully facilitates secure data rewriting and ensures accountability, all while maintaining an acceptable level of additional time overhead.

With the rapid increase in user demand for business traffic, the deployment and migration of virtual service function chains (VSFC) considering network load balancing has become a fascinating research topic. In this paper, we delve into the cascading failures in virtual network function (VNF) migration and design a topology-awareness fault-tolerant migration mechanism for VSFC. Firstly, we design a measurement model for the evolution of network nodes under cascading faults, which provides an evaluation method for node importance. Secondly, we present aTopology andResourceAware VNF fault-tolerant migration framework (TRA) under cascading faults. Finally, we investigate a topology-awareness energy consumption optimization algorithm based on the cascading failure. The proposed algorithm can reduce network energy consumption and alleviate network link congestion while reasonably migrating virtual machines to meet system performance requirements. Through extensive simulation experiments and real testbed evaluations,TRAreduced migration time and average latency by 28.6% and 24.7% compared with the average performance of VVi, TeaVisor, and MiOvnm, respectively. In addition,TRAhas improved revenue to expense ratio and load balancing index by 19.6% and 26.5%, respectively.

The cyclical and continuous working characteristics ofInternet of Things(IoT) devices make a large amount of the same or similar data, which can significantly consume storage space. To solve this problem, various secure data deduplication schemes have been proposed. However, existing deduplication schemes only perform deduplication based on data similarity, ignoring the internal connection among devices, making the existing schemes not directly applicable to parallel and distributed scenarios like IoT. Furthermore, since secure data deduplication leads to multiple users sharing same encryption key, which may lead to security issues. To this end, we propose a device relationship-based IoT data deduplication scheme that fully considers the IoT data characteristics and devices internal connections. Specifically, we propose a device relationship prediction approach, which can obtain device collaborative relationships by clustering the topology of their communication graph, and classifies the data types based on device relationships to achieve data deduplication with different security levels. Then, we design a similarity-preserving encryption algorithm, so that the security level of encryption key is determined by the data type, ensuring the security of the deduplicated data. In addition, two different data deduplication methods, identical deduplication and similar deduplication, have been designed to meet the privacy requirement of different data types, improving the efficiency of deduplication while ensuring data privacy as much as possible. We evaluate the performance of our scheme using five real datasets, and the results show that our scheme has favorable results in terms of both deduplication performance and computational cost.

The design of interconnection network topologies significantly impacts the performance and reliability of parallel systems. Enhanced incremental scalability enables networks to expand with reduced hardware overhead. In practice, rather than always adding many nodes at once, a small number of nodes are occasionally added as needed. However, existing topologies struggle to achieve effective incremental scalability. To address this, we propose the incremental scalability exchanged hypercube (ISEH), a novel interconnection network for parallel computing. The significant advantages of ISEH include improved incremental scalability and interconnection flexibility, while maintaining low interconnection complexity. Its diameter remains unchanged as the network size increases linearly and does not exceed the diameter of the exchanged hypercube. First, we present the topological properties of ISEH, including isomorphism, incremental scalability, and diameter. Next, we design an efficient communication method for ISEH to ensure low communication overhead. To support reliable communication, we design algorithms to construct disjoint paths between any two distinct nodes. Furthermore, based on generalized exchangedX-cubes, we propose the incremental scalability generalized exchangedX-cubes, offering better incremental scalability. Finally, we compare the performance of ISEH with other interconnection networks and evaluate the proposed algorithms. The results demonstrate that ISEH achieves a favorable balance among incremental scalability, diameter, and flexibility compared to existing networks.

The rise of the industrial Internet of Things (IIoT) has enhanced industrial processes through interconnected devices and data exchange, but it also introduces significant security vulnerabilities, such as malware attacks, which threaten system reliability and availability. To address this challenge, we extend the traditional reliability block diagram (RBD) method by integrating stochastic games to evaluate the security and availability of IIoT systems. Our approach constructs a comprehensive Markov transition matrix using additional node states, enabling detailed simulations of malware spread in IIoT networks. By modeling the interactions between malware and IIoT systems through stochastic games, we propose an innovative reinforcement learning algorithm named evaluation-driven Q-learning (EDQL) to solve these complex scenarios. This novel application of EDQL in the realm of availability evaluation is a significant contribution, providing a rare integration of game theory into this field. We also derive the availability of individual IIoT nodes using reliability theory and integrate these insights into the RBD framework. Experimental results demonstrate that the EDQL algorithm significantly outperforms traditional reinforcement learning methods in malware reward. Furthermore, our method effectively evaluates common IIoT topologies and offers practical deployment recommendations, highlighting its practical impact and significance in enhancing IIoT system security and availability.

In the application of the Internet of Things (IoT), existing cloud edge collaboration technologies face the problem of poor coordination of heterogeneous resources. In this article, we proposeCFEMC, which is a novelCloud-Fog-EdgeMulti-layerCollaboration resource scheduling framework for IoT. First, we design a collaborative resource scheduling framework based on semi-distributed artificial intelligence. It can achieve collaborative optimization of cloud/edge computing resource allocation under the constraints of high reliability and low latency. Second, we present a workflow applications scheduling strategy based on the proposed collaborative resource scheduling framework. This can solve the problem of unstable computing performance and transmission bandwidth during the scheduling process. Finally, the extensive and real data supported simulation results show thatCFEMChas advantages in terms of energy consumption, delay and throughput compared with other benchmark strategies. Against CEC Hu et al. 2023 and PSO Zeng et al. 2022, the average throughput increases by 16.37% and 24.21%, and the total queuing delay decreases by 54.23% and 58.12%, respectively.

Multi-hop cooperative task offloading (MCTO) allows resource-constrained edge clouds to collaborate and assist each other in completing computation-intensive tasks, such as training machine learning models through device-edge-cloud (DEC) computing. However, internal fake service attacks can pose a threat to the security and reliability of MCTO in DEC computing. In this paper, we propose a trust model based on a directed acyclic graph (DAG) and Proof-of-Work (PoW) to safeguard tasks against potential attacks. The edge node selection for task offloading involves two key steps: offloading confirmation and trust-based node selection. To mitigate the unreliability caused by internal fake service attacks during cooperative offloading, we propose a multi-hop offloading node selection algorithm based on the soft actor-critic (SAC) coalition. This algorithm helps identify trustworthy nodes for constructing secure offloading paths. Our experimental results demonstrate that the proposed algorithm effectively counters internal fake service attacks and significantly reduces cooperative offloading latency compared to existing leading approaches.

Internet of Things (IoT) systems are susceptible to compromise due to malware propagation, leading to the data breach and information theft. In this paper, we propose a proactive deception-oriented hypergame-theoretic malware propagation-mitigation (DHMPM) model between IoT nodes and edge devices under asymmetric information in edge intelligence (EI)-enabled IoT systems. We then explore malware-propagated deceptive defense strategies based on deep reinforcement learning. Specifically, IoT nodes and edge devices continually adjust their strategies based on obtained utilities under beliefs perceived by uncertainties from the game environment and system dynamics. Built upon the proposed game DHMPM, we next apply spiking neural networks (SNNs) into deep Q-network to form hypergame-theoretic deep spiking Q-network (HGDSQN), practically converging to the optimal malware-propagated deceptive defense strategy in EI-enabled IoT systems. Such SNNs can simulate biological brains with the pulse communication mechanism and break through the bottleneck of temporal processing in traditional models with deep neural networks, realizing intelligent decision-making and real-time malware defense. We eventually perform experimental simulations that assess the effect of attack arrival probability and learning rate on the optimal learning strategy selection, demonstrating the effectiveness of the proposed HGDSQN algorithm.

With the growing need to comply with privacy regulations and respond to user data deletion requests, integrating machine unlearning into IoT-based federated learning has become imperative. This article introduces an innovative framework that melds blockchain with federated learning, ensuring an immutable record of unlearning requests and actions. Our approach not only bolsters the trustworthiness and integrity of the federated learning model but also adeptly addresses efficiency and security challenges typical in IoT environments. Key contributions include a certification mechanism for the unlearning process, enhancement of data security and privacy, and optimization of data management. Experimental results on MNIST and CIFAR-10 datasets demonstrate the effectiveness of our approach, achieving 0% accuracy for unlearned classes while maintaining 77.74% and 42.65% overall model accuracy for MNIST and CIFAR-10, respectively. Our time complexity analysis shows that the blockchain integration introduces only 2 seconds of overhead per epoch, highlighting the practicality of our solution for IoT applications.

Since the waveforms of the signals received at adjacent times are different due to varying modulated signals, it is still a key challenge on how to describe temporal correlations precisely in compressive spectrum sensing (CSS). In this paper, we propose a higher-order cumulant-assisted CSS (HOC-CSS) algorithm, where the semantic correlation of adjacent received signals in higher-order domain is exploited. Firstly, the fourth-order cumulants of the reconstructed signals in the previous time are calculated. Secondly, the fourth-order cumulants are integrated with$\ell _{p}$-norm minimization to estimate the current spectrum signals. Furthermore, a two-branch CSS scheme is proposed to eliminate the effect causing by adjacent signals under different modulations. The HOC-CSS is performed in one branch and a noninformative CSS algorithm is performed in another branch, thereby two different reconstructed signals are obtained. The Euclidean distances between the measurements and the projection of reconstructed signals are calculated. The final output is the estimated signal with smaller Euclidean distance. The experimental results demonstrate that the proposed algorithm has more than 8% and 10% detection performance improvement at false alarm probability 0.1, under compressive ratio of 0.4 and 0.5 respectively, compared with the state-of-the-art CSS schemes, e.g., HMCPI-CSS, Weighted$\ell _{1}$-CSS, AR-IRLS, and IP-IRLS.

The outstanding performance of machine learning models across various fields enables them to be used in sensitive and high-risk activities such as healthcare, automated driving, and security services. However, the explainability of their outputs and the safety of their operation are major concerns. Although Explainable AI (XAI) can enhance the interpretability of AI models, further research is necessary to evaluate its effectiveness in explaining adversarial attacks. The use of XAI techniques and the rise of adversarial attacks are important issues related to the explainability and security of deep learning, respectively. Furthermore, the relation between the explainability and safety of deep learning, such as the vulnerability of the XAI explanation to an adversarial attack, is key to unlocking these concerns. In this paper, we use the Saliency Map as an XAI technique to explain the behavior of the Fast Gradient Sign Method (FGSM) adversarial attack on the ResNet model, and show the vulnerability related to such an explanation with respect to the attack. Extensive experiments show that as the severity of FGSM attack on the ResNet model increases, the Saliency Map gradually fails, exposing its potential vulnerability.

Backdoor attacks significantly threaten machine learning-based vision systems. Existing detection methods typically require clean data from a similar distribution as the dataset under inspection, limiting practical deployment. This work proposes a Mutation-Based Method (MBM) for detecting and filtering backdoored samples in image training dataset, without referencing any external clean data. MBM aims at distinguishing backdoored and benign samples distribution via their distinct stability in feature space under certain data augmentations. Firstly, MBM applies multiple data augmentation techniques, generating mutated versions of each sample to ‘deactivate’ potential triggers while maintaining natural semantics not heavily distorted. Secondly, MBM measures how sample features diverge after mutating from its origin as poison score, which we call ‘Feature Stability’. Thirdly, by analyzing extreme scores within each class, MBM effectively identifies the backdoored class, and isolates samples not from backdoored class as clean data. Finally, a benign distribution is fit to benchmark against backdoored samples from backdoored class. We validated MBM on the CIFAR-10 dataset, achieving a true positive rate above 95% and a false positive rate below 0.2% for all defense settings. Our results confirm MBM’s efficacy without reliance on external clean data.

A Digital Twin (DT) is a software agent of a physical entity in virtual space, transcending the limitations imposed by physical constraints to enable intelligent services. The emergence of the Internet of Digital Twins (IoDT), by connecting DTs as a network, provides a reliable solution for streamlining the sharing of extensive real-time data among DTs. However, ensuring the security and efficiency of highly sensitive DT data remains a significant challenge in the distributed and dynamic IoDT storage environment. In this paper, we propose a comprehensive scheme that integrates blockchain, Distributed Hash Table (DHT), and Attribute-Based Encryption (ABE) to establish an IoDT data-sharing system. First, we introduce an ABE scheme to achieve data confidentiality and fine-grained access control. By implementing a key generation protocol, we address key escrow concerns in ABE. Second, to ensure bidirectional data confidentiality in data sharing without revealing the identities of DTs, the DHT, and blockchain are jointly applied to achieve resource registration and discovery. Through an XOR mapping operation mechanism, DT data are distributed across different DHT network nodes, reducing DT nodes’ storage consumption and effectively addressing the scalability issue of IoDT storage. Finally, we provide performance evaluations to demonstrate the reliability and efficiency of our proposed scheme. Our scheme strikes a balance between efficiency and security, exhibiting efficient performance in IoDT data sharing compared to other ABE and blockchain schemes.

Federated Learning (FL) is a distributed machine learning (ML) approach that enables multiple edge nodes to collaboratively train ML models by sharing model parameters, thus addressing privacy concerns. However, in highly distributed, dynamic, and volatile FL environments, the global model is vulnerable to various corruptions. For instance, edge nodes might falsely claim that the received global model is incomplete, or the channel that transmits the global model is untrustworthy. Effectively verifying the integrity of the global model poses a critical challenge. To tackle this issue, we introduce a method for verifying model integrity called Federated learning global Model Integrity Verification (FMIV). It leverages Merkle Hash Tree (MHT) to generate integrity proofs of the global model during verification. To improve security, we integrate random security codes during proof generation. FMIV is capable of verifying the global model updated by the central server and shared with untrusted edge nodes, while efficiently identifying the edge node caching the incomplete global model. Furthermore, we conduct theoretical analysis and extensive experiments to validate the performance of FMIV. Compared to the two state-of-the-art approaches, FMIV consistently exhibits a notable improvement in verification efficiency and effectiveness in detecting model corruption.

Recent years have witnessed increasing privacy concerns towards machine learning. To protect privacy in machine learning, federated learning has been proposed as a decentralized privacy-preserving framework where clients upload the parameters rather than private data. However, training a fair federated learning model in heterogeneous environments is still challenging. First, heterogeneous data distributions lead the global model fail to show high accuracy on all distributions. Second, the federated learning training process exposes and exacerbates potential biases in heterogeneous training data. Third, the local bias of each client can be propagated through parameter sharing, biasing the global model. In this work, we propose a two-stage fairness-aware federated learning framework (HeteroFair) to achieve fairness under heterogeneous data distributions. Initially, we introduce the fairness constraint to the loss function and propose a local adaptive weighting algorithm to adjust the proportion of the fairness constraint, achieving fair training in heterogeneous environments. Then, we present a fairness-aware aggregation reweighting algorithm that reduces the mismatch between local and global fairness to achieve fair federated learning. Extensive evaluation results demonstrate the effectiveness of our proposed framework in achieving fairness and high accuracy under het-eroaeneous data distributions.

Due to the complexity of industrial environments, such as construction sites and production workshops, the objects to be detected are easily occluded and perceived as small objects, which poses certain challenges for object detection. To ensure a safe industrial environment, this study adopts YOLOv5 as the basic framework and integrates the depth-to-space convolution module to improve the model’s ability to extract feature information of small targets. Second, the global attention mechanism is incorporated into the network to enhance the global interaction information, reducing the feature information loss, and improving the model performance. Finally, to alleviate the contradiction between classification and regression tasks in object detection, the YOLOv5 head is replaced with a decoupled head to achieve better classification and accelerate model convergence. To improve data diversity and enhance model robustness, we augmented the open-source safety helmet wearing dataset (SHWD) and smoking behavior detection dataset (SBDD). We test the performance of the proposed model (SGD-YOLOv5) on the large and small object detection dataset (SODA-D) and VisDrone2021-DET datasets. Furthermore, its ability to detect small objects was also evaluated. Experiments show that our model outperforms all baseline models on SHWD and SBDD. Compared to the TPH-YOLOv5 on the SODA-D dataset, AP and Recall achieve improvements of 18.3% to 19.2% and 11.9% to 13.3% respectively. On the Visdrone 2021-DET dataset, [email protected] achieved an improvement from 35.45% to 35.70% compared to the state-of-the-art model YOLO-Drone.

This work presents a highly effective strategy for attacking image captioning models through the use of prompt engineering. The objective of this approach is to deliberately guiding the output of LLMs and introduce dynamic noise into the original clean image captions, causing them to be categorized as a different class. Consequently, when the image captioning model is fine-tuned using adversarial captions, it will deteriorate and produce inaccurate captions for clean photos. The novelty of this attack is that it does not require the attacker to perform any model training and only require to prompt the LLMs to generate only a small amount of captions for the attack to be effective. Comprehensive experiments using GPT-3.5 indicate that with only 100 captions created by LLMs with malicious intent can significantly worsen picture captioning model performance by up to over 50% in BLEU metric and over 25% in ROUGE-L and METEOR scores.

Note:This paper contains many malicious contents generated by LLMs. Jailbreak can cause large language models (LLMs) to violate moral and ethical guidelines, generating harmful text and images. However, the existing jailbreaks are high-cost (e.g., fine-tuning LLMs) and underperform when facing some specific jailbreak tasks (e.g., generating pornographic and bloody texts), which has too many limitations for reflecting the threat of jailbreak on LLMs’ security. To fill this gap, in this paper, we propose a black-box jailbreak with a higher attack success rate and lower cost, called Dictionary Jailbreak with Multi-round Induction (DJMI). First, in DJMI, we create a simple custom language dictionary based on specific jailbreak tasks to be performed and send it to the LLM. Then, we use the custom language from the dictionary to construct malicious prompts, instructing the LLM to respond in the custom language as much as possible. Generally, in the first round, the LLM will provide a neutral response (such as translating the malicious prompts) or refuse to answer. To counteract this, we need to emphasize that the prompts comply with ethical guidelines and repeat the prompts using the custom language. The process can be conducted with multiple rounds to guide LLM in outputting harmful content. During the attack process, the only attack cost is the transmission cost of the prompts through the official interactive interface, without any other costs of LLM fine-tuning, algorithm design and prompt generation. Thus, DJMI is a low-cost jailbreak method. We verified DJMI on multiple mainstream LLMs across various jailbreak tasks. Experiments show that compared to existing black-box jailbreaks, DJMI achieves a higher attack success rate (> 80% on average) across specific jailbreak tasks with various risk levels. Additionally, DJMI enables the LLM to provide highly detailed execution steps for harmful behaviors within a session (such as specifying proportions of ingredients, synthetic chemical formulas, experimental conditions, and other detailed information in illicit drug production), further highlighting the severity of jailbreak attacks.

Wearable Internet of Things (IoT) devices are easily compromised by malware due to their security vulnerabilities. The bots infected by malware may continue to infect healthy neighbor devices through wireless communication technology in mobile wearable IoT networks (WIoT). All bots form a botnet which eventually leads to a series of malicious attacks. Therefore, it is necessary to predict the dynamic malware propagation path between wearable devices, which can help provide target immunization measures on devices to prevent the formation of botnets. In this article, we capture the local interaction and spatial–temporal propagation behavior of malware utilizing the individual-based cellular automata (CA) model. First, taking into account the mobility of walking users carrying wearable devices in the actual WIoT, we present a human mobility model called Gauss–Markov truncated Levy walk (GM-TLW) to describe the movement patterns of mobile users. Second, based on the moving coordinates of all wearable devices obtained from the GM-TLW mobility model, we leverage the improved CA propagation model to study the time evolution of the number of bots and the spreading spatial distribution of malware. We compare our propagation model with the differential equation model and traditional CA model, and analyze the impact of various parameters on the dynamics of botnet formation using numerical simulations. Finally, detailed simulation results show that the GM-TLW model is more suitable for realistic human mobility scenarios. In addition, the proposed CA-based model is more precise than the differential equation model to modeling the malware propagation and provides a basis for defenders to adopt the optimal malware control strategies.

In the Internet of Things (IoT) scenario, the device diversity and data sparsity present a significant challenge for malicious traffic detection, notably the “small sample problem” where insufficient data hampers the performance of the deep learning methods that depend on large volumes of labeled data for training. Transfer learning (TL) has the capability to transfer knowledge from a label-rich but heterogeneous domain to a label-sparse domain, making it a powerful tool for addressing challenges in IoT malicious traffic detection. To address these challenges, we introduce the EMTD-SSC model, a novel enhanced malicious traffic detection model that leverages TL under small sample conditions in IoT environments. Initially, our approach includes a comprehensive labeled data set that merges a small-scale IoT intrusion detection domain with the traditional intrusion detection domain to enrich semantic information transfer from the source to target domains. The EMTD-SSC model employs dual residual convolutional autoencoders for robust feature extraction and transfer, incorporating skip connections to expedite the model convergence and minimize information loss. Furthermore, to optimize transfer efficiency, we minimize the multilayer multi kernel maximum mean discrepancy (MLMK-MMD) across corresponding network layers, facilitating effective domain adaptation. Through unsupervised training and subsequent fine tuning on the target domain data, the model significantly enhances anomaly detection capabilities. Extensive experiments on the two well-known public data sets demonstrate that the EMTD-SSC model’s effectiveness, achieving an impressive 94.8% accuracy in the binary classification tasks.

Recently, with the development of automatic driving, high-precision localization has become indispensable for safety. The inertial measurement unit (IMU) integrated with ultra-wideband (UWB) technology has been thought as a promising scheme by the company Humatics for trains. Generally, the spaces of transportation are extended along a long path with flexural tunnels and sideway obstacles around, where non-line-of-sight (NLoS) propagation happens frequently. However, a large interval deployment of UWB anchors along such narrow path with NLoS interference will dramatically decrease localization performance. Hence, we propose a robust algorithm to mitigate the NLoS interference. Firstly, a joint LoS/NLoS detection and mitigation algorithm is proposed to improve the ranging accuracy of mobile target with UWB tags under mixed LoS/NLoS interference. Secondly, we improve the conventional Kalman Filter (KF) algorithm with forward-backward propagation to integrate temporal correlations further. Finally, a comprehensive fusion localization scheme with ranging error mitigation and improved KF is proposed. The simulation results show that, regarding the static localization, the proposed joint LoS/NLoS detection and mitigation algorithm outperforms five other typical NLoS elimination algorithms, and achieves 48.2%, 62.4%, 45.5%, 69.9% and 50.2% gains in terms of root mean square error (RMSE), under NLoS scenarios with a mean ranging error of 0.5 m. Regarding the dynamic localization, compared with two other typical KF localization algorithms, the proposed localization scheme achieves 20.9% and 14.5% localization accuracy improvements in LoS scenarios, 45.7% and 47.2% improvements in mixed LoS/NLoS scenarios, respectively. Furthermore, the effectiveness of the proposed scheme is also verified in our testing platform.

In recent years, the utilization of fog cloud-based Internet of Things (IoT) applications has been steadily rising due to the exponential growth of data produced by interconnected smart devices. However, cloud providers who are responsible for these IoT applications face two critical problems: 1) how to protect the system from untrusted users and 2) how to allocate processing units to meet the demands with acceptable costs. The fog–cloud blockchain system (FCB), proposed in past research, provides a perfect solution for the former question by integrating Blockchain’s security qualities into the fog–cloud paradigm. In this article, we address the latter question by proposing an improved version of the life-choice-based optimization algorithm (ILCO) to solve the task scheduling for Bag-of-Task applications in the FCB system. Task scheduling is one of the most prominent problems in resource allocation. Our proposed algorithm not only increases the convergence speed but also maintains diversity better, optimizing the FCB’s power, latency, and cost. Under a single-objective problem setting, ILCO outperforms LCO and similar state-of-the-art methods by achieving better results for FCB’s latency and power consumption.

Distributed Denial-of-Service (DDoS) attacks pose a serious threat to the stability and security of edge intelligence devices. To solve this issue, we first describe the cost of edge intelligence environments in detail and introduce the mean-field term, edge intelligence repair speed, and DDoS attack intensity, which provide a theoretical basis for the subsequent model construction. Secondly, based on Hamilton-Jacobi-Bellman (HJB) backward and Fokker-Planck-Kolmogorov (FPK) forward equations, a mean-field model is proposed. The optimal DDoS defense policy is then solved considering the strongest DDoS attack intensity and the best edge intelligence device repair speed. Further, the focus is turned to the interaction between DDoS attackers and the edge intelligence environment. We give the update rule of the mean-field term and construct a mean-field game (MFG) model with a value function. Finally, we propose a multi-agent deep reinforcement learning algorithm called Mean-Field Games with Dueling Double Deep Q-learning Network (MFGD3QN) to solve the optimal DDoS defense policy problem under the MFG model. In the experiment, we compare MFGD3QN with several benchmark algorithms and verify the superiority of the MFGD3QN algorithm in an edge intelligence environment. We also carry out experiments on different parameters of the MFGD3QN algorithm, lower visibility conditions of the edge intelligence environment, and different repair speeds of edge intelligence devices, which verify the robustness and feasibility of the algorithm.

Massive volumes of malware spread incidents continue to occur frequently across the Internet of Things (IoT). Owing to its self-learning and adaptive capability, artificial intelligence (AI) can provide assistance for automatically converging to an optimal strategy. By merging AI into edge computing, we consider an edge intelligence-enabled IoT (EIIoT) environment and provide a stochastic learning strategy for suppressing the spread of IoT malware. In particular, we introduce stochastic game theory to symbolise the whole process of the confrontation between IoT malware and edge nodes. Built upon the theoretical framework to demonstrate the specific spread-suppression architecture, we apply the improved Deep Q-Network algorithms including DDQMS, D2QMS and D3QMS that can deduce the optimal EIIoT malware spread-suppression strategy with better performance. Through experiments, we investigate the influence of related parameters on learning strategy selection, recommending the optimal parameters setting of automated EIIoT malware spread-suppression. We also compare the performance of the proposed three DQN-improved algorithms.

Existing mobile edge computing (MEC) systems are facing the challenges of limited resources and highly dynamic network environments. How to allocate resources to maintain the efficiency and timeliness of data and tasks is still an open issue. To address this problem, we propose a novel framework for UAV-assisted MEC systems using federated multi-agent reinforcement learning. First, we formulate a joint optimization problem as a multi-agent Markov decision process by jointly minimizing the average age of information and maximizing the number of recent tasks. Second, we design a novel scheduling algorithm for online collaborative resources by adopting multiple agents to learn and make decisions in accordance with the overall interests through federal learning. Finally, an experience replay mechanism for the internal experience pool is introduced to further improve learning efficiency. Experimental results show that our proposed algorithm is superior to the recent typical reinforcement learning-based algorithms. It not only has higher efficiency in task processing and data freshness, but also has more stable performance and adaptability across diverse experimental conditions.

The rapid growth and escalating complexity of the Internet of Things (IoT) necessitate meticulous attention to ensure efficient and secure transactions among various autonomous components. To address this critical issue, this study proposes the integration of multiagent systems (MASs) and blockchain technology within the IoT domain. Uniquely, our approach employs smart contracts to manage exchanges between autonomous entities, thereby offering enhanced security, transparency, and reliability. The study introduces a set of innovative algorithms that regulate agent activities, such as creating blocks, sharing information, and conducting rating processes. Additionally, it provides a detailed analysis of their privacy and security aspects. Compared to traditional multiagent frameworks, empirical evidence demonstrates significant improvements in efficiency, adaptability, and scalability. This scholarly effort lays a robust foundation for further investigations into applying blockchain to enhance MASs, potentially paving the way for more sophisticated and context-specific strategies across various IoT fields.

Federated learning (FL) FL has emerged as a promising privacy-preserving machine-learning technology, enabling multiple clients to collaboratively train a global model without sharing raw data. With the increasing adoption of FL in Internet of Things (IoT) scenarios, concerns about security and privacy have become critical. In particular, gradient inversion attacks and poisoning attacks pose significant threats to the integrity and effectiveness of the global model. In response, we propose a comprehensive blockchain-based defense mechanism that effectively protects FL systems from such attacks. We develop a novel combination of techniques, including public blockchain level protection and private blockchain level protection, which work in tandem to prevent attackers from reconstructing figures using the obtained gradients. This unique combination of methods provides a robust defense against gradient inversion attacks in FL IoT scenarios. We conduct extensive experiments to validate the effectiveness of our proposed approach against gradient inversion and poisoning attacks. Our results demonstrate improved accuracy and stable convergence of training loss under poisoning attacks, indicating that our method can be applied to a wide range of FL IoT scenarios, enhancing both the security and privacy of distributed machine-learning systems.

Implementing federated learning within the Internet of Everything (IoE) framework poses substantial computational challenges, stemming from extensive client involvement, which can lead to escalated training expenses and diminished convergence rates. While many studies have investigated the combination of federated learning and blockchain networks, the integration of public and private chains to enhance federated learning performance remains largely unexplored. In this study, we introduce an innovative methodology that unifies public and private chains to mitigate clients’ computational demands while preserving data privacy and security, demonstrating compatibility within the IoE milieu and yielding favorable outcomes. To facilitate secure model migration and expedite training without incurring excessive computation costs, we delineate a blockchain-anchored model migration scheme tailored for resource-limited IoT infrastructures, establishing a private chain mechanism to incentivize companies possessing multiple devices or clients to prioritize model training. Employing blockchain technology guarantees trustworthiness in model migration, precluding the disclosure of devices’ confidential data. Overall, our innovative method provides an effective solution that improves the accuracy, privacy, and security of federated learning while reducing clients’ computational burdens within the context of the Internet of Everything (IoE).

Opportunistic computation offloading is an effective way to improve the computing performance of Industrial Internet of Things (IIoT) devices. However, as more and more computing tasks are being offloaded to mobile-edge computing (MEC) servers for processing, it can lead to IIoT privacy and security issues, such as personal usage habits. In this paper, we aim to design a Lyapunov-based privacy-aware framework that defines the amount of IIoT user privacy and designs a “reduced amount of privacy” mechanism. We first define the cumulative privacy amount for each IIoT user and trigger the privacy protection mechanism when the cumulative privacy amount exceeds the set privacy threshold. The offloading data generated by the IIoT user is then transferred to local processing, and finally, the cumulative privacy amount of the IIoT user is reduced. This model ensures that the cumulative privacy of all IIoT users remains stable. We further combine the advantages of Lyapunov optimization and actor-critic networks to address the problem of how to make the model learn the optimal policy and maintain the minimum energy consumption in the long run. Especially, this framework integrates model-based optimization and model-free actor-critic networks to handle the offloading problem with very low computational complexity, and Lyapunov optimization ensures that this framework minimizes energy consumption while stabilizing the data queue. It is demonstrated through experimental simulation results that the proposed scheme can maintain data queue stability and minimize energy consumption under strict security.

The escalating growth of the Internet of Things (IoT) has intensified the demand for dependable and efficient communication networks to accommodate the massive data volumes produced by interconnected devices. Satellite networks have emerged as a promising alternative, particularly in remote and underserved regions where terrestrial communication infrastructures are inadequate. Nevertheless, guaranteeing secure uplink communications in satellite-based IoT networks is a daunting task due to similar satellite channels and limited resources at IoT nodes. In this article, we explore the potential of unmanned aerial vehicle (UAV) to improve the secrecy performance of uplink transmissions in satellite-supported IoT networks. Specifically, we first introduce a framework for UAV-aided secure uplink communications, presuming a secure UAV-to-satellite connection. To mitigate the risks of ground eavesdroppers intercepting uplink transmissions, we develop a max–min secrecy rate optimization problem with uplink power constraints. To address this nonconvex problem, a streamlined two-stage optimization approach is proposed. In the inner stage, we combine uplink power allocation and UAV beamforming and propose a successive convex approximation (SCA)-based joint optimization algorithm to address them. In the outer stage, we propose a synergized bisection and coordinate descent algorithm to optimize UAV positioning. Convergence is attained by alternating iterations between these two stages. Particularly, the secrecy fairness among IoT users is reached by solving the max–min problem. Additionally, we offer a complexity analysis of the proposed algorithm and validate the efficacy of the presented approach through comprehensive simulation results.

The Industrial Internet of Things (IIoT) has experienced rapid growth in recent years, with an increasing number of interconnected devices, thereby expanding the attack surface. Effectively detecting intrusions is crucial for safeguarding IIoT systems from malicious attacks. However, due to the dynamic and complex nature of the IIoT environment, designing an intrusion detection strategy that balances accuracy and efficiency remains a significant challenge. In this paper, we propose a novel intrusion detection strategy based on stochastic games and deep reinforcement learning (DRL) for detecting attacks effectively while balancing detection accuracy and efficiency in the IIoT. We model the interaction between attackers and detectors as dynamic adversarial stochastic games with incomplete information, theoretically analyze Nash equilibria, and construct a node-based simulation of interconnected infrastructure within the IIoT. We then propose a novel algorithm DDQN-LP combining Double Deep Q-Networks with “lazy penalty” to determine optimal strategies and encourage agents to promptly conclude the game to reduce overhead. Furthermore, we identify different optimal hyperparameters for training our DRL agents and evaluate their efficacy both theoretically and empirically. We compare our proposed algorithm with other reinforcement learning algorithms, and simulations demonstrate our approach has better performance with a higher detection rate as well as lower consumption.

Industrial Internet of Things (IIoT) has brought a lot of convenience for the industrial world to digitization, automation and intelligence, but it inevitably introduces inherent cyber security risks, resulting in an issue that traditional intrusion detection techniques are no longer sufficient for IIoT environments. To solve this issue, we propose an open-set solution called DC-IDS for IIoT based on deep reinforcement learning. In this solution, the open-set recognition problem in intrusion detection is modeled as a discrete-time Markov decision process, and Deep Q-Network (DQN) is employed to solve it. Meanwhile, a Conditional Variational Auto-Encoder is introduced to the value network in DQN. Therefore, the open-set recognition problem in intrusion detection is divided into two subproblems, namely known traffic fine-grained classification problem and unknown attacks recognition problem. We use DQN to solve the known traffic fine-grained classification problem. Since the reconstruction error of known traffic is generally smaller than the reconstruction error of unknown attacks, we use reconstruction error to recognize unknown attacks. Experiments on IIoT dataset TON-IoT demonstrate the effectiveness of DC-IDS model, which achieves better performance in terms of the recognition rate of unknown attacks as well as the stability of the model compared to previous proposed methods.

The coexistence of sixth-generation (6G) and other wireless networks in the large-scale environment is challenged by the problem of the efficient sharing of the unlicensed spectrum band, as well as the fair and reliable access of heterogeneous users. Blockchain with the advantages of trustworthiness and decentralization is supposed to be a potential solution. An intelligent two-layer blockchain framework is developed, consisting of an upper layer administrator blockchain (UAB) and multiple lower layer spectrum allocation blockchains (LSB). The LSB is an adjustable local blockchain to form a fast and lightweight consensus on spectrum allocation, and the UAB controls the adaptation for regional dynamic Quality-of-Service requirements. Moreover, the theoretical model is proposed to analyze the effect of different communication delays of wireless blockchain networks on the block confirmation delay of the directed acyclic graph (DAG) blockchain. Considering the wireless blockchain node must compete for scarce spectrum resources to broadcast transactions, we analyze the optimal spectrum allocation ratio for an LSB through mathematical modeling. Furthermore, the deep Q network (DQN) is used by UAB to form a consensus on the adjustment strategy for the blockchain model of LSB. Test results demonstrate that the proposed intelligent blockchain model can significantly improve spectrum efficiency and adapt to the dynamic environment.

The rapid advancement of wireless communication and artificial intelligence (AI) has led to a plethora of emerging applications that require exceptional connectivity, minimal latency, and substantial computing resources. The widespread adoption of cloud-edge intelligence is propelling the development of future networks capable of supporting intelligent computing. Mobile edge computing (MEC) technology facilitates the movement of computing resources and storage to the network’s edge, enabling cost-effective offloading of computational tasks for related applications which needs for reduced latency and improved energy efficiency. However, the offloading efficiency is hindered by limitations of wireless transmission capacity. This paper aims to address this issue by integrating reconfigurable intelligent surfaces (RISs) into a cell-free network within an intelligent cloud-edge system. The core idea is to strategically deploy passive RISs around base stations (BSs) to reconstruct the transmission channel and improve the corresponding capacity. Subsequently, we formulate an optimal problem involving joint beamforming for RISs and BSs, which is characterized by non-convexity and complexity. To tackle this challenge, we employ an alternating optimization scheme to ensure the effectiveness of joint beamforming. In particular, deep reinforcement learning (DRL) is leveraged to reduce the computational complexity involved in optimizing task offloading. Additionally, Lyapunov optimization is utilized to model the latency queue and improve the learning efficiency of the offloading framework. We conduct comprehensive evaluations on the wireless system’s capacity, average latency, and energy consumption, considering the integration of RIS with the DRL offloading framework. Experimental results demonstrate that our proposed scheme achieves superior efficiency and robustness.

Routing, working with scheduling cooperatively to ensure the deterministic low-latency communication, plays a critical role in Time-Sensitive Networking (TSN). The multi-level spatial correlation among network nodes may greatly impact the forwarding paths of flows and is yet to be well investigated to improve the paths selecting process. In this letter, hypergraph is first introduced to model the time-sensitive networks, which can capture group features and higher-dimensional relationships for nodes. Then, the forwarding process for flows on a hyperedge and the end-to-end hyperpaths of flows are analyzed and studied. Finally, a hypergraph-based routing algorithm is designed to provide efficient routing decisions for time-sensitive flows. Simulation results show that the proposed routing algorithm can enhance the efficiency of paths planning and provide satisfiable global schedule for time-sensitive networks.

3D tori are significant interconnection architectures in building supercomputers and parallel computing systems. Due to the rapid growth of edge faults and the crucial role of path structures in large-scale distributed systems, fault-tolerant path embedding and correlated issues have drawn widespread researches. However, existing path embedding methods are based on traditional fault models, allowing all faults to be near the same node, so they usually only focus on theoretical proof and generate linear fault-tolerance related to dimension$n$. In order to improve the fault-tolerance of 3D torus, we first propose a novel conditional fault model called the Locally Faulty Block model (LFB model). On the basis of this model, the Hamiltonian paths with large-scale edge defects in torus are investigated. After that, we construct an Hamiltonian path embedding algorithm HP-LFB into torus with$O(N)$under the LFB model, where$N$is the number of nodes in torus. Furthermore, we present an adaptive routing algorithm HoeFA, which is based on the method of distance vector to limit the use of virtual channels (VCs). We also make a comparison with state-of-the-art schemes, indicating that our scheme enhance other comprehensive results. The experiment indicated that HP-LFB can sustain the dynamic degradation of the batting average of establishing Hamiltonian paths, with the added faulty edges exceeding fault-tolerance.

Model prediction based on machine learning is provided as a service in cloud environments, but how to verify that the model prediction service is entirely conducted becomes a critical challenge. Although zero-knowledge proof techniques potentially solve the integrity verification problem when applied to the prediction integrity of massive privacy-preserving Convolutional Neural Networks (CNNs), the significant proof burden results in low practicality. In this research, we present psvCNN (parallel splitting zero-knowledge technique for integrity verification). The psvCNN scheme effectively improves the utilization of computational resources in CNN prediction integrity, proving by an independent splitting design. Through a convolutional kernel-based model splitting design and an underlying zero-knowledge succinct non-interactive knowledge argument, our psvCNN develops parallelizable zero-knowledge proof circuits for CNN prediction. Furthermore, psvCNN presents an updated Freivalds algorithm for a faster integrity verification process. Experiments show that psvCNN is practical and efficient in terms of proof time and storage, generating a prediction integrity proof with a proof size of 1.2MB in 7.65s for the structurally complicated CNN model VGG16. psvCNN is 3765 times faster than the latest zk-SNARK-based non-interactive method vCNN, and 12 times faster than the latest sumcheck-based interactive technique zkCNN in terms of proving time.

The use of social network to model the evolution of credit scores of networked enterprises is still a challenging task. This article develops an opinion dynamics model of the evolution of credit scores of enterprises in a social network. Firstly, based on the number of potential cooperated enterprises and the initial credit scores, the leader and follower enterprises are identified. Then, taking into consideration the cooperated benefit and discrimination cost, the cooperated utility between any two enterprises is calculated, which is used to compute the weights that one enterprise assigns to other enterprises. An opinion dynamics model on the evolution of credit scores of enterprises, inspired on the classical Friedkin–Johnsen’s social network model, is developed. Some desirable properties of the proposed opinion dynamics model are theoretically stated and proved. Finally, a numerical example is provided to illustrate the feasibility of the proposed opinion dynamics model, while a simulation analysis to investigate the joint influences of the connection probabilities and the network structure on the evolution of credit scores of enterprises is reported.

Federated learning (FL) trains a model over multiple datasets by collecting the local models rather than raw data, which can help facilitate distributed data analysis in many real-world applications. Since the model parameters can leak information about the training datasets, it is necessary to preserve the privacy of the FL participants’ local models. Furthermore, FL is vulnerable to poisoning attacks which can significantly decrease the model utility. To settle the above issues, we propose a privacy-preserving and Byzantine-robust FL scheme$\Pi _{\text{P2Brofl}}$that maintains robustness in the presence of poisoning attacks and preserves the privacy of local models simultaneously. Specifically,$\Pi _{\text{P2Brofl}}$leverages three-party computation (3 PC) to securely achieve a Byzantine-robust aggregation method. To improve the efficiency of privacy-preserving local model selection and aggregation, we propose a maliciously secure top-$k$protocol$\Pi _{\text{top}-k}$that has low communication overhead. Moreover, we present an efficient maliciously secure shuffling protocol$\Pi _{\text{shuffle}}$since secure shuffling is necessary for our secure top-$k$protocol. The security proof of the scheme is given and experiments on real-world datasets are conducted in this paper. When the proportion of Byzantine participants is 50%, the error rate of the model only increases by 1.05% while it increases by 23.78% without using our protection.

With the development of the Industrial Internet of Things (IIoT), the scale of IIoT data is rapidly increasing, bringing significant challenges to existing data management systems. To tackle this issue, we propose a similarity-based secure deduplication for IIoT cloud management system, which can effectively balance the security and availability of IIoT data and minimize the storage cost. Concretely, we propose a similarity-based secure deduplication algorithm for IIoT (IIoT-SBSD) by designing a similarity-preserving tag (IIoT-Simhash). This algorithm can perform similarity deduplication over ciphertexts, thus reducing storage space while ensuring data security. Besides, we construct a parallelizable edge-based deduplication framework in which similarity comparison and deduplication operations are performed directly by edge nodes, significantly alleviating the transmission pressure. Additionally, we propose similarity-based proofs of ownership, S-PoWs, to mitigate the impact of data deduplication on the user's access to the IIoT data. Experimental results show that our method significantly reduces storage space and transmission bandwidth without increasing the computation burden.

The flexibility of 5G-NGNs makes them an ideal infrastructure for supporting mission-critical IoT applications that require low latency and high bandwidth. However, due to the rapid proliferation and the integration of IoTs with 5 G, the threat surface has considerably expanded. Hence the security of IoT devices is a big concern. Unfortunately, IoT devices have limited resources, and the traditional security approaches (authentication and intrusion detection approaches) of cryptography do not work effectively on 5G-IoT ecosystems. Motivated from this, we leverage the distinctive RF (Radio Frequency) fingerprinting signatures of IoT devices and used them to train a Deep learning model, Mahalanobis Distance theory in addition to the Chi-square distribution theory, to authenticate the IoT nodes. Under robust scenarios we have tested the approach shows detection accuracy (99.35%) as well as significant amount of reduction in model's training time as these two metrics are one of the primary key performance indicators (KPIs). In order to evaluate the effectiveness of the proposed method in real-time scenarios, we tested the proposed solution with a real RF dataset and the OSM-MANO 5 G platform. The model underwent formal verification using the Tamarin Prover tool, and the proposal was also compared with recent research works.

With pervasive IoT networking, traffic-analysis-based IoT fingerprinting techniques have been well researched. For example, by integrating blockchain technology and device fingerprinting, authentication of devices connected to a network can be achieved. Though the primary motivations are identifying vulnerabilities and implementing access control, the techniques could be exploited to trace IoT users’ privacy. We propose a traffic morphing scheme to protect IoT devices from being identified by fingerprinting models. The scheme mainly consists of a morphing policy learning algorithm, a rewarding model, and a time-series-based feature estimation algorithm. Backed by the timely rewarding model, a learning agent produces an optimal policy that perturbs the target fingerprinting model while preserving the original traffic function. The estimation algorithm predicts the feature vectors of unfinished flows to enable live traffic morphing. The scheme's advantage is that it requires minimal knowledge of the fingerprinting model and supports live morphing. Experimental results show that over 81% of the IoT devices become unidentifiable, and the scheme degrades the average F1 score of mainstream fingerprinting models from 0.996 to 0.526. For certain devices and target models, the scheme even reaches 100% effectiveness.

Model inversion attacks can reconstruct the training samples of victim deep learning models. The existing efforts heavily rely on auxiliary information of the target samples (prior target information) to achieve their adversarial goals. However, prior target information is hard to obtain in practice. In this paper, we explore the effect of class information in model inversion attacks to reduce the reliance of prior target information. Our contributions on class information exploitation are two-fold. Firstly, we propose a supervised inversion model, Supervised Model Inversion (SMI). The proposed inversion model learns pixel-level features and data-to-class features from the rounded-outputs of the victim model and labeled auxiliary dataset. Secondly, we leverage victim model's rounded-outputs to guide the optimization of reconstructing inversion samples after trained inversion model. Our experimental results show that inversion samples reconstructed by SMI are more visually plausible with more details, comparing to the three representative model inversion attacks. We further perform an extensive study on various auxiliary dataset settings. It is found that the class combination in the auxiliary dataset rather than the number of classes that determines the quality of inversion samples. The ground-truth labels can improve the qualities of inversion samples but not essential to inversion attacks.

Although Federated Learning (FL) prevents the exposure of original data samples when collaboratively training machine learning models among decentralized clients, it has been revealed that vanilla FL is still susceptible to adversarial attacks if model parameters are leaked to malicious attackers. To enhance the protection level of FL, Differential Private Federated Learning (DPFL) has been proposed in recent years. DPFL injects zero-mean noises randomly generated by differential private (DP) mechanisms on local model parameters before they are disclosed. Nevertheless, DP noises can significantly deteriorate model utility jeopardizing the practicality of DPFL. In this paper, we are among the first to explore how to improve the model utility of DPFL by tuning the number of local iterations (LIs) on DPFL clients. Our work shows that such a local iteration tuning approach can well mitigate the adverse influence of DP noises on the final model utility. Formally, we derive the sensitivity (a measure of the maximum change of the output given two adjacent inputs) with respect to the number of LIs conducted on DPFL clients for the Laplace mechanism, and the aggregated variances of Laplace noises at the server side. We further conduct convergence rate analysis to quantify the influence of the Laplace noises on the final model accuracy and determine how to optimally set the number of LIs. Finally, to verify our theoretical findings, we perform extensive experiments using three real-world datasets, namely, Lending Club, MNIST and Fashion-MNIST. The results not only corroborate our analysis, but also demonstrate that our approach significantly improves the practicality of DPFL.

In recent years, there is an explosive growth in multimodal learning. Image captioning, a classical multimodal task, has demonstrated promising applications and attracted extensive research attention. However, recent studies have shown that image caption models are vulnerable to some security threats such as backdoor attacks. Existing backdoor attacks against image captioning typically pair a trigger either with a predefined sentence or a single word as the targeted output, yet they are unrelated to the image content, making them easily noticeable as anomalies by humans. In this paper, we present a novel method to craft targeted backdoor attacks against image caption models, which are designed to be stealthier than prior attacks. Specifically, our method first learns a special trigger by leveraging universal perturbation techniques for object detection, then places the learned trigger in the center of some specific source object and modifies the corresponding object name in the output caption to a predefined target name. During the prediction phase, the caption produced by the backdoored model for input images with the trigger can accurately convey the semantic information of the rest of the whole image, while incorrectly recognizing the source object as the predefined target. Extensive experiments demonstrate that our approach can achieve a high attack success rate while having a negligible impact on model clean performance. In addition, we show our method is stealthy in that the produced backdoor samples are indistinguishable from clean samples in both image and text domains, which can successfully bypass existing backdoor defenses, highlighting the need for better defensive mechanisms against such stealthy backdoor attacks.

Insider threats represent a significant challenge in both corporate and governmental sectors. Most existing supervised learning based detection methods that rely on transforming user behavior into sequential data do not fully utilize structural information and require extensive labeled data. This reliance poses a challenge due to the scarcity of labeled data in real-world scenarios, leading to a few-shot learning situation. To address these limitations, we propose a novel Graph modularized-based Meta-learning Framework for Insider Threat Detection, named GMFITD. Specifically, GMFITD utilizes a structural reconstruction mechanism that combines a graph-based autoencoder with an attention mechanism to explore structural information and infer potential relationships between users. Additionally, we employ a graph prototype construction method coupling episodic meta-learning principle (MAML) to compute representative embeddings for few-shot learning scenarios. By leveraging MAML, the proposed method can capture prior knowledge of insider threat classification by training on similar few-shot learning tasks with few labeled samples. We further enhance the resilience of GMFITD to adversarial attacks through an edge importance estimation mechanism, which assigns higher weights to relevant edges. Extensive experiments demonstrate that our proposed GMFITD outperforms state-of-the-art methods in insider threat detection, achieving higher accuracy with fewer labeled samples and resisting adversarial attacks.

Malware propagation in IoT (Internet of Things) systems can lead to data leakages, financial losses, and other serious consequences. To solve this issue, we propose a new active IoT malware propagation defence work. Specifically, aided by stochastic games, we express the process of cyber conflicts between IoT system nodes and edge devices considering malware propagation in edge intelligence-enabled IoT. Here, IoT system nodes and edge devices choose their own strategies and receive the corresponding rewards determined by the current state and strategy. After that, the game randomly moves to the next stage according to the distribution of probabilities and the participants’ strategies until reaching the fixed Nash equilibrium point. Following a theoretical analysis, we design and implement SGD3QN (Stochastic Games and Dueling Double Deep Q-networks)—a novel algorithm to receive the optimal strategy for mitigating IoT malware propagataion in practice. Here, the Dueling Double Deep Q-networks are acted as an end-to-end decision control system, in which IoT malware propagataion environment is used as the input to obtain the failure or success experience to update the network parameters, followed by making the optimal decision output. Afterwards, we perform experimental simulations that probe the influence of batch size and replay memory size on the optimal IoT malware propagation defense strategy selection and prove the ascendancy of the proposed SGD3QN-aided decision-making algorithm.

Machine unlearning enables data owners to remove the contribution of their specified samples from trained models. However, existing methods fail to strike an optimal balance between erasure effectiveness and model utility preservation. Previous studies focused on removing the impact of user-specified data from the model as much as possible to implement unlearning. These methods usually result in significant model utility degradation, commonly called catastrophic unlearning. To address the issue, we systematically consider machine unlearning and formulate it as a two-objective optimization problem that involves forgetting the erased data and retaining the previously learned knowledge, highlighting accuracy preservation during the unlearning process. We propose an unlearning method called representation-forgetting unlearning with parameter self-sharing (RFU-SS) to achieve the two-objective unlearning goal. Firstly, we design a representation-forgetting unlearning (RFU) method that aims to remove the contribution of specified samples from a trained representation by minimizing the mutual information between the representation and the erased data. The representation is learned using the information bottleneck (IB) method. RFU is tailored to the IB structure models for ease of introduction. Secondly, we customize a parameter self-sharing structural optimization method for RFU (i.e., RFU-SS) to simultaneously optimize the forgetting and retention objectives to find the optimal balance. Extensive experimental results demonstrate a significant effectiveness improvement of RFU-SS over the state-of-the-art methods. RFU-SS almost eliminates catastrophic unlearning, reducing model accuracy degradation from over 6% to less than 0.2% on the MNIST dataset with an even better removal effect. The source code is available athttps://github.com/wwq5-code/RFU-SS.git.

Industrial Internet platforms (IIP) can provide many intelligent services based on the industrial big data stored in clouds. However, the vulnerability of cloud storage can cause data corruption, demanding verifying its integrity. Unfortunately, existing cloud storage verification approaches cannot be directly applied to IIP, since they pose heavy computational burdens on the edge side. In this work, we propose an efficient and dynamic storage verification scheme Edasvic for cloud storage in the IIP. We adopt the polynomial commitment to build an efficient homomorphic authenticator, and further design an authenticator accumulator, which can be efficiently generated with limited computational overheads. In addition, we integrate the dynamic information into the authenticator accumulator to support data dynamics. The security of Edasvic is analyzed under the random oracle model. We conduct extensive experiments to evaluate the performance of Edasvic and compare it with the state-of-the-art approaches. Experimental results affirm that Edasvic is superior to existing solutions in terms of computational efficiency.

In light of the growing emphasis on the right to be forgotten of graph data, machine unlearning has been extended to unlearn the graph structures’ knowledge from graph neural networks (GNNs), namely, structure unlearning. Whereas the complex dependencies in graph data, structure unlearning is intrinsically prone to imbalanced performance between the objectives of knowledge forgetting and model utility maintenance. Nevertheless, most existing methods fall short in addressing the two objectives in tandem and developing balanced solutions. In this paper, we propose imbalanced Structure Unlearning Mitigation using MultI-objective OpTimization (SUMMIT), which aims to develop balanced solutions regarding both knowledge forgetting and model utility maintenance effects. Corresponding to the two aspects, we first construct two tailored objectives that specifically address the challenges inherent in structure unlearning. Specifically, for the forgetting objective, we introduce a higher-order forgetting enhancement strategy aimed at mitigating the adverse effects of GNN oversmoothing on node decoupling. For the remembering objective, we adhere to the principle of ideal unlearning and propose to minimize the distributional distance between the node embeddings developed by unlearned and well-trained GNNs. Considering the potential competitive relationship between the two objectives during the optimization process, we present an adaptive two-objective balancer based on multi-objective optimization to reconcile the two objectives and strike a balance between them. We conduct comprehensive experiments to evaluate the efficacy of SUMMIT on three representative GNNs and four datasets, and compare the performance of SUMMIT with its ablation variants and a cadre of baselines. We demonstrate the superiority of SUMMIT in its ability to yield optimal and balanced solutions, addressing both the facets of knowledge forgetting and model utility maintenance.

Recent studies have demonstrated that backdoor attacks can cause a significant security threat to federated learning. Existing defense methods mainly focus on detecting or eliminating the backdoor patterns after the model is backdoored. However, these methods either cause model performance degradation or heavily rely on impractical assumptions, such as labeled clean data, which exhibit limited effectiveness in federated learning. To this end, we proposeFLPurifier, a novel backdoor defense method in federated learning that can effectively purify the possible backdoor attributes before federated aggregation. Specifically,FLPurifiersplits a complete model into a feature extractor and classifier, in which the extractor is trained in a decoupled contrastive manner to break the strong correlation between trigger features and the target label. Compared with existing backdoor mitigation methods,FLPurifierdoesn’t rely on impractical assumptions since it can effectively purify the backdoor effects in the training process rather than an already trained model. Moreover, to decrease the negative impact of backdoored classifiers and improve global model accuracy, we further design an adaptive classifier aggregation strategy to dynamically adjust the weight coefficients. Extensive experimental evaluations on six benchmark datasets demonstrate thatFLPurifieris effective against known backdoor attacks in federated learning with negligible performance degradation and outperforms the state-of-the-art defense methods.

Machine learning (ML) has been widely adopted for Android malware detection to deal with serious threats brought by explosive malware attacks. However, it has been recently proven that ML-based detection systems exhibit inherent vulnerabilities to evasion attacks, which inject adversarial perturbations into a malicious app to hide its malicious behaviors and evade detection. To date, researchers have not found effective solutions for this critical problem. Although there are some similar works in the image classification field, most of those ideas cannot be borrowed due to the significant differences between images and Android apps. In this paper, we exploit Moving Target Defense (MTD) to continually change the attack surface of the protected detector and create uncertainty on the attacker side. We thus propose a novel Android malware detection framework named MTDroid, which fully leverages a seamless blend of dynamicity, diversity, and heterogeneity to mitigate the impact of evasion attacks. To this end, we develop a dynamic model pool to decrease the exposure time of a single classifier, by building and rebuilding multiple heterogeneous models with distinct data. We then generate diversified variant models to provide defensive measures against various attacks, and further improve robustness through ensemble learning. Specifically, we propose a two-stage selection algorithm to optimize the ensemble learning process, and design a hybrid update strategy to refresh the framework dynamically. The experimental results show that MTDroid significantly enhances the robustness against a wide range of attacks and outperforms the state-of-the-art methods upon three popular practical datasets.

Intelligent Connected Vehicles (ICVs) need to obtain real-time traffic data from nearby ICVs or remote content providers to ensure safe driving. However, providers are hesitant to share their data due to privacy and benefits concerns. To ensure privacy while improving efficiency of obtaining data, we proposed smart contract-based data sharing among ICVs, and content delivery between ICVs and remote content provider. To solve low willingness to vehicles due to untrustworthy third-party platforms, we use smart contracts to implement access control during data upload and transaction. Then, we propose a one-to-many sharing model based on Stackelberg game to model the interaction between consumers and owners. Consumers adjust their reward strategies with the owners’ optimal strategies to maximize its utility, thus obtaining the nash equilibrium solution. To provide reliable quality of service (QoS) and security guarantee for content delivery, smart contracts regulate the delivery process, facilitating automatic execution under specific conditions. Transaction records audited and stored on blockchain enhance transparency and trustworthiness. Utilizing a delivery utility model that considers benefits, costs, and mining profits, proposed quantum particle swarm optimization (QPSO) algorithm is used to find the optimal solution. We built an EdgeChain testbed, and used BDD-100K dataset to evaluate the performance in utility, access delay, etc. Compared to CTM and MFPA, proposed data sharing algorithm achieves maximum consumer utility. Compared to LRU, PCCM and MARL, when content is 400, proposed content delivery algorithm reduces average access delay by 30.88%, 18.92% and 4.86%, and reduce backhaul load by 50.04%, 47.23% and 3.16%.

Vehicle-based mobile crowdsensing has gained widespread attention due to its low cost and efficient data collection mode. One common method to improve the accuracy of sensing data in this context is truth discovery. However, the emergence of privacy leakage and data misuse has reduced users’ motivation to participate in sensing tasks. Meanwhile, existing solutions for privacy-preserving truth discovery generally suffer from low computational efficiency and frequent interactions between users and servers. Hence, this paper proposes a novel privacy-preserving truth discovery scheme based on secure multi-party computation. For the purpose of high efficiency and strong privacy protection, we utilize the Secret Sharing method to securely decompose data and construct a Secure Multi-party Computation protocol to compute the ground truth. In addition, the weight value generated by truth discovery is employed as a quantitative data quality indicator that dynamically adjusts the user’s rewards and constructs a data quality-driven incentive mechanism. Finally, we demonstrate the high performance of our method through a detailed analysis, showing its effectiveness even in scenarios with numerous users.

Most existing methods for visual domain adaptation need to convert high-order tensors into one-order high-dimensional vectors through naive vectorization operations. However, they not only destroy the internal spatial structure within the original high-order tensors, but also result in exponentially increasing model parameters. To address these problems, we propose a novel method for visual domain adaptation by representing tensorial features in tensor-train subspace in this paper. Specifically, we firstly provide a theoretical deduction by constructing a tensor-train subspace and proving its linearity and left-orthogonality. Secondly, to extract common tensorial features between source and target domains, we formulate the visual domain adaptation problem into an optimization problem that models the aforementioned common tensor-train subspace between two domains, as well as their corresponding projections. Thirdly, we design a tensor-train subspace representation algorithm (TTSR) to solve the multiple variables optimization problem by optimizing its sub-problems iteratively, so as to process high-order tensorial features. Finally, we evaluate the performance of our proposed TTSR algorithm by conducting extensive experiments on three popular public datasets. The experimental results demonstrate that the TTSR algorithm can improve the classification accuracy of unlabeled target domain than that of baseline algorithms.

This paper examines integrating jamming and secondary signals for covert communications in cognitive radio networks (CRNs), aiming to enhance covertness by using jamming and secondary signals in an overlay cooperative CRN. The scenario involves a primary base station (PBS) transmitting to a primary user (PU), with a secondary user transmitter (SU-Tx) acting as a cooperative jammer to obscure the message from a malevolent secondary user named “Willie.” During idle intervals on the primary channel, the SU-Tx opportunistically accesses it to transmit secondary signals, reinforcing the covert communication of primary signals. The study quantifies the detection error probability (DEP) experienced by Willie, considering perfect and statistical channel state information (CSI) scenarios. In the perfect CSI scenario, optimization has two phases. Phase I aims to maximize the signals-to-interference-plus-noise ratio (SINR) of the PU, subject to the warden DEP exceeding a specified threshold. Phase II uses an iterative search algorithm to optimize beamforming vectors, enhancing SINR. In the statistical CSI scenario, the goal is to maximize effective transmission throughput (ETT), measuring the information transmitted from PBS to PU under covert constraints. Numerical results validate the theoretical analysis.

Smart devices (SDs) used in the Industrial Internet of Things can generate computational tasks for processing the data generated during production. However, due to the limited processing power of SDs, it is necessary to transfer these computational tasks to more powerful devices for processing. To this end, we propose a Mobile Edge Computing (MEC) system based on a Software Defined Network (SDN) for SDs to offload their computational tasks. This MEC system includes multiple MEC servers to handle numerous SDs, which leads to load-balancing challenges among these servers. To tackle this problem, we develop a computational offloading model based on mean-field game theory and introduce a mean-field game-based load-balancing algorithm (MFGLB), which reduces processing latency and facilitates task scheduling through Multi-Agent Deep Reinforcement Learning. Each SD in the MEC system is considered a participant in the mean-field game, simplifying the complex stochastic game into a more manageable dual-agent game. We then prove the existence of Nash Equilibrium for this mean-field game. To evaluate the effectiveness of our MFGLB algorithm, we compare its performance with traditional load-balancing algorithms and a stochastic game-based load-balancing algorithm. Our experimental results demonstrate the superiority of MFGLB in reducing processing latency and addressing load imbalances.

With the development of wireless communications, the 6G network is evolving toward dynamics, complexity, and integration. The mobility of nodes and intermittently of links lead to frequent variations in the network topology. When constructing the topology model, the reliability of topology is not only affected by the physical properties of wireless links but also related to the evolution process of nodes and links states, which is indispensable for improving the accuracy of the topology model. In this paper, we propose the evolution model based on dynamic bond percolation to characterize the reliable topology evolution. Firstly, key factors that cause the network topology changes are analyzed, integrating the characteristics of the node mobility and link channel conditions. Especially signal interference, buffer of nodes, and link availability are modeled for wireless link states. Then, the interactions between adjacent links are formulated by an extended Dynamic Bond Percolation (DBP) model to obtain the topology state transition matrix, which can accurately depict the change of link connection. Based on the quantitative analysis of wireless link states, Markov chain and master equation are employed to build the Dynamic Topology Evolution (DTE) model. Meanwhile, the network topology prediction problem is transformed into a linear system solution problem to obtain the steady-state network topology based on the DTE algorithm. Finally, the results suggest that utilizing the DTE model can significantly improve the accuracy of topology prediction and overall network performance.

The emergence of mobile edge computing (MEC) imposes an unprecedented pressure on privacy protection, although it helps the improvement of computation performance including energy consumption and computation delay by computation offloading. To this end, we concern about the privacy protection in the MEC system with a curious edge server. We present a deep reinforcement learning (DRL)-driven computation offloading strategy designed to concurrently optimize privacy protection and computation cost. We investigate the potential privacy breaches resulting from offloading patterns, propose an attack model of privacy theft, and correspondingly define an analytical measure to assess privacy protection levels. In pursuit of an ideal computation offloading approach, we propose an algorithm, SAC-PP, which integrates actor-critic, off-policy, and maximum entropy to improve the efficiency of learning processes. We explore the sensitivity of SAC-PP to hyperparameters and the results demonstrate its stability, which facilitates application and deployment in real environments. The relationship between privacy protection and computation cost is analyzed with different reward factors. Compared with benchmarks, the empirical results from simulations illustrate that the proposed computation offloading approach exhibits enhanced learning speed and overall performance.

Recent advances in Cloud Computing have shown great power in enhancing intelligent devices to support various applications. Nevertheless, conventional Cloud Computing fails to keep up with the ever-advancing requirements of efficient task execution, mainly resulting from its drawbacks in communication delay. To this end, multi-layer cloud computing with local, edge, and remote data centers has gained high interest yet remains challenging because of the inherent complexity of cross-layer orchestration. In particular, with more participants involved, it is nontrivial to achieve customized service provision while guaranteeing system stability. Hence, we address the workload scheduling issue in the multi-layer cloud paradigm in this paper, with task offloading and service reconfiguration considered jointly. We first formulate it as a stochastic optimization problem, where statistical service requirements are imposed on queue lengths. Then, we divide the original optimization into three individual low-complex sub-problems with optimal solutions provided. To improve system performance, we introduce a request-rejecting mechanism that augments our approach with delay-optimality. Theoretical analysis confirms that our approaches can guarantee system stability and are asymptotically optimal within a small gap from the optimum. Finally, we validate the efficiency of our approaches through extensive simulation results in performance guarantees and customized workload scheduling.

A key feature of federated learning (FL) is to preserve the data privacy of end users. However, there still exist potential privacy leakage in exchanging gradients under FL. As a result, recent research often explores the differential privacy (DP) approaches to add noises to the computing results to address privacy concerns with low overheads, which however degrade the model performance. In this paper, we strike the balance of data privacy and efficiency by utilizing the pervasive social connections between users. Specifically, we propose SCFL, a novel Social-aware Clustered Federated Learning scheme, where mutually trusted individuals can freely form a social cluster and aggregate their raw model updates (e.g., gradients) inside each cluster before uploading to the cloud for global aggregation. By mixing model updates in a social group, adversaries can only eavesdrop the social-layer combined results, but not the privacy of individuals. As such, SCFL considerably enhances model utility without sacrificing privacy in a low-cost and highly feasible manner. We unfold the design of SCFL in three steps. i) Stable social cluster formation. Considering users’ heterogeneous training samples and data distributions, we formulate the optimal social cluster formation problem as a federation game and devise a fair revenue allocation mechanism to resist free-riders. ii) Differentiated trust-privacy mapping. For the clusters with low mutual trust, we design a customizable privacy preservation mechanism to adaptively sanitize participants’ model updates depending on social trust degrees. iii) Distributed convergence. A distributed two-sided matching algorithm is devised to attain an optimized disjoint partition with Nash-stable convergence. Experiments on Facebook network and MNIST/CIFAR-10 datasets validate that our SCFL can effectively enhance learning utility, improve user payoff, and enforce customizable privacy protection.

In recent years, there has been significant advancement in the field of model watermarking techniques. However, the protection of image-processing neural networks remains a challenge, with only a limited number of methods being developed. The objective of these techniques is to embed a watermark in the output images of the target generative network, so that the watermark signal can be detected in the output of a surrogate model obtained through model extraction attacks. This promising technique, however, has certain limits. Analysis of the frequency domain reveals that the watermark signal is mainly concealed in the high-frequency components of the output. Thus, we propose an overwriting attack that involves forging another watermark in the output of the generative network. The experimental results demonstrate the efficacy of this attack in sabotaging existing watermarking schemes for image-processing networks with an almost 100% success rate. To counter this attack, we propose an adversarial framework for the watermarking network. The framework incorporates a specially-designed adversarial training step, where the watermarking network is trained to defend against the overwriting network, thereby enhancing its robustness. Additionally, we observe an overfitting phenomenon in the existing watermarking method, which can render it ineffective. To address this issue, we modify the training process to eliminate the overfitting problem.

Since the introduction of Bitcoin in 2008, blockchain technology has found widespread applications across various domains. While blockchain offers convenience and immense research value, it also raises privacy and security concerns among users and society at large. Notably, numerous studies have demonstrated the vulnerability of blockchain anonymity. Existing solutions based on bloom filters and SGX(Software Guard Extensions) may safeguard users' access patterns but remain susceptible to novel attacks, including protocol-level and side-channel attacks. To address these issues, we propose a Tree-based Oblivious Random Access Protocol (Tree-ORAP) that not only provides access pattern protection in privacy-preserving blockchain systems but also preserves the original blockchain performance. Furthermore, we design a Tree-ORAP State Version Controller to manage state synchronization across nodes in a multi-client blockchain network. We also analyze the system's security and implement a Tree-ORAP prototype, conducting a series of experiments to demonstrate its efficiency and technical feasibility. In summary, our protocol offers enhanced protection for blockchain systems against a wider range of attacks compared to previous methods, all while maintaining superior security performance and equal or better efficiency.

The booming of learning-based Artificial Intelligence (AI) enables the integration of Big Data and emerging computing architectures, which facilitate the Edge-AI-as-a-Service (EAaaS) in the edge-cloud continuum. To meet the emerging demands, such as privacy preservation and autonomy, blockchain-enabled federated learning (B-FL) is proposed, which further provides decentralized processing, data falsification avoidance, and learning model reliability. However, synchronous global aggregation, which is deployed in most existing B-FL paradigms, is dragging down the performances due to the data and computing resources heterogeneity of diverse edge devices. In addition, the restricted resources of edge devices pose further challenges in executing learning tasks and blockchain-based consensus simultaneously. To solve these issues, we propose a blockchained dual-asynchronous federated learning (BAFL-DT) service model for EAaaS in the digital twin empowered edge-cloud continuum. In BAFL-DT, federated learning services are run on local edge devices, while the global aggregation is achieved by the consensus process of digital twins implemented in the cloud. Besides, dual-asynchronous FL allows both local training and global aggregation to be performed in an asynchronous manner, which is uniquely enabled by the proposed paradigm. Extensive evaluations of real-world datasets testify to the superior performances of EAaaS by improving accuracy and efficiency.

The publicly released machine learning (ML) models are susceptible to malicious attacks (e.g., gradient leakage attacks), which may expose sensitive training data of data-sharing platforms to untrusted third-parties. To preserve the privacy of training data, differential privacy (DP) is exploited to limit the amount of leaked privacy with a predefined budget, which in fact is a non-recoverable resource. Considering DP, allocating privacy budgets to ML queries is a non-trivial but crucial problem because a certain amount of non-recoverable privacy budget will be consumed if a datablock is assigned to a query once. Meanwhile, both datablocks and ML queries are continuously generated, which further complicates the problem. Most existing works simply relied on greedy-based algorithms to make myopic allocation decisions, far away from the optimal decision. In this paper, we propose a novelHistory-awarePrivacyBudgetAllocation (HPBA) algorithm for data-sharing platforms to address the above challenges. Different from existing works, HPBA leverages historical query records to approximate global ML query patterns so as to overcome the drawback of shortsighted greedy-based algorithms. Moreover, the performance of HPBA is theoretically guaranteed by competitive analysis. A lightweight version called S-HPBA is proposed to further reduce computation overhead by using fewer historical records. Experimental results demonstrate that, compared to the state-of-the-art baselines, HPBA and S-HPBA improve the average performance by 32.8% and 16.2% in terms of model accuracy, respectively.

Blockchain technology, known for its decentralized and immutable nature, serves as the foundation for various applications. As a prominent application of blockchain, decentralized storage is powered by blockchain technology and is expected to provide a reliable and cost-effective alternative to traditional centralized storage. A major challenge in blockchain-powered decentralized storage is how to guarantee the quality of storage services in decentralized storage nodes (DSNs). Storage auditing can ensure the integrity and security of the stored data. Unfortunately, it incurs additional computational costs for data owners and extra storage overheads for DSNs, which thereby cannot be directly applied to decentralized storage networks consisting of nodes with various computation and storage capacity. In this article, we overcome these problems and minimize additional burdens in storage auditing. We propose EDCOMA, a computation and storage efficient auditing scheme for blockchain-based decentralized storage, in which a double compression method is designed to compress data authenticators using both data and polynomial commitment. To prevent replay attacks on double compression launched by DSNs, we introduce zero knowledge proof and design a compression arithmetic circuit to guarantee the execution of compression operations in DSNs. We analyze the security of EDCOMA under the random oracle model and conduct extensive experiments to evaluate the performance of EDCOMA. Experimental results affirm that EDCOMA outperforms state-of-the-art approaches in both computational and storage efficiency.

Compressive spectrum sensing (CSS) is regarded as one of the promising techniques to detect wideband spectrum holes. How to formulate an accurate prior knowledge in CSS is still an important and open topic. In this paper, we propose a novel CSS algorithm with cascade forward-backward propagation and prior knowledge refining. Firstly, the wideband signal reconstruction model is formulated as a$\ell _{p}$-norm ($0 < p < 1$) nonconvex optimization problem where the temporal correlation between continuous observed spectrum signals is exploited to improve the reconstruction accuracy of the current spectrum signal, and such nonconvex optimization problem is solved by iteratively reweighted least squares algorithm. Then, the multiple reconstructed versions of the current spectrum signal obtained by forward-backward propagation are fused into a prior. Finally, such fused prior is modified adaptively in each iteration, and thus significantly improves the accuracy of spectrum decision and speeds up the convergence. The simulation results show that compared with other CSS schemes based on weighted$\ell _{1}$minimization,$\ell _{1}$-$\ell _{1}$minimization,$\ell _{1}$-$\ell _{2}$minimization, and maximizing correlation, the proposed scheme has more than 7% and 10% detection performance improvement when the probability of false alarm is 10%, under received signal-to-noise ratio as −10 dB and −5 dB, respectively.

Since hybrid beamforming (HBF) can approach the performance of fully-digital beamforming (FDBF) with much lower hardware complexity, we investigate the HBF design for beam-hopping (BH) low earth orbit (LEO) satellite communications (SatComs). Aiming at maximizing the sum-rate of totally illuminated beam positions during the whole BH period, we consider joint beamforming and illumination pattern design subject to the HBF constraints and sum-rate requirements. To address the non-convexity of the HBF constraints, we temporarily replace the HBF constraints with the FDBF constraints. Then we propose an FDBF and illumination pattern random search (FDBF-IPRS) scheme to optimize illumination patterns and fully-digital beamformers using constrained random search and fractional programming methods. To further reduce the computational complexity, we propose an FDBF and illumination pattern alternating optimization (FDBF-IPAO) scheme, where we relax the integer illumination pattern to continuous variables and after finishing all the iterations we quantize the continuous variables into integer ones. Based on the fully-digital beamformers designed by the FDBF-IPRS or FDBF-IPAO scheme, we propose an HBF alternating minimization algorithm to design the hybrid beamformers. Simulation results show that the proposed schemes can achieve satisfactory sum-rate performance for BH LEO SatComs.

As the right to be forgotten has been legislated worldwide, many studies attempt to design machine unlearning mechanisms to enable data erasure from a trained model. Existing machine unlearning studies focus on centralized learning, where the server can access all users’ data. However, in a popular scenario, federated learning (FL), the server cannot access users’ training data. In this paper, we investigate the problem of machine unlearning in FL. We formalize a federated unlearning problem and propose a bayesian federated unlearning (BFU) approach to implement unlearning for a trained FL model without sharing raw data with the server. Specifically, we first introduce an unlearning rate in BFU to balance the trade-off between forgetting the erased data and remembering the original global model, making it adaptive to different unlearning tasks. Then, to mitigate accuracy degradation caused by unlearning, we propose BFU with parameter self-sharing (BFU-SS). BFU-SS considers data erasure and maintaining learning accuracy as two tasks and optimizes them together during unlearning. Extensive comparisons between our methods and the state-of-art federated unlearning method demonstrate the superiority of our proposed realizations.

As graphs are getting larger and larger, federated graph learning (FGL) is increasingly adopted, which can train graph neural networks (GNNs) on distributed graph data. However, the privacy of graph data in FGL systems is an inevitable concern due to multi-party participation. Recent studies indicated that the gradient leakage of trained GNN can be used to infer private graph data information utilizing model inversion attacks (MIA). Moreover, the central server can legitimately access the local GNN gradients, which makes MIA difficult to counter if the attacker is at the central server. In this paper, we first identify a realistic crowdsourcing-based FGL scenario where MIA from the central server towards clients’ subgraph structures is a nonnegligible threat. Then, we propose a defense scheme, Subgraph-Out-of-Subgraph (SOS), to mitigate such MIA and meanwhile, maintain the prediction accuracy. We leverage the information bottleneck (IB) principle to extract task-relevant subgraphs out of the clients’ original subgraphs. The extracted IB-subgraphs are used for local GNN training and the local model updates will have less information about the original subgraphs, which renders the MIA harder to infer the original subgraph structure. Particularly, we devise a novel neural network-powered approach to overcome the intractability of graph data’s mutual information estimation in IB optimization. Additionally, we design a subgraph generation algorithm for finally yielding reasonable IB-subgraphs from the optimization results. Extensive experiments demonstrate the efficacy of the proposed scheme, the FGL system trained on IB-subgraphs is more robust against MIA attacks with minuscule accuracy loss.

Federated learning (FL) is an emerging paradigm using a parameter server (PS) to coordinate multiple decentralized clients for training a common model without exposing their raw data. Despite its amazing capability in preserving data privacy, FL confronts two significant challenges that have not been sufficiently addressed by existing works, which are: 1) heterogeneous data distributed on clients given that the PS cannot alter data locations. 2) limited computation resources of FL clients who may conduct model training with mobile devices. To tackle these challenges, we propose a novel federated one-shot transferring of intermediate features (FedOTF) algorithm. More specifically, FedOTF consists of two stages: feature extraction and model reconstruction. To overcome challenge 1, clients in stage 1 only collaboratively train a small model in a federated manner, with the objective to extract features. In stage 2, intermediate features (generated by the small model trained by stage 1) are transferred from clients to the PS so that a large model can be trained to overcome challenge 2. In addition, the design of FedOTF is robust, which can flexibly diminish the number of communication rounds of stage 1 when network capacity is limited, and reduce the amount of exposed features when privacy is concerned. To verify the superiority of FedOTF, we conduct comprehensive experiments with real datasets. The experiment results demonstrate that FedOTF can significantly improve the model utility of FL because a better model can be finally obtained by the PS without incurring heavy computational load on clients. Besides, we conduct robustness evaluation of FedOTF which can achieve stable performance when varying network capacity and privacy requirement.

Federated learning (FL) requires clients to train constituted models based on their local datasets. Clients usually directly train local models using their entire datasets without distinguishing which information of data is task-relevant or irrelevant. Task-irrelevant information does not contribute to the learning task but exposes additional privacy information to adversaries. Studies have shown that unintended information leakage from gradients during FL iterations threatens clients' privacy. Researchers applied differential privacy (DP) to protect clients' gradients, but it does not help to reduce task-irrelevant information from the gradients. In this paper, we propose a compressive privacy federated learning (CP-FL) scheme to protect the task-irrelevant information from gradient leakage attacks. In CP-FL, clients train a local compressive model according to the global task. The local compressive model constructs a new representation, which extracts task-relevant and removes task-irrelevant information from clients' data. Since the global model is updated based on the compressed representation that eliminates the task-irrelevant information, it can effectively prevent adversaries from inferring those property values from the uploaded gradients. Moreover, with the help of a powerful local compressive model that sanitizes the challenging data into a low-dimension space representation, CP-FL can use a small global model instead of a sizeable one, significantly reducing communication. Both theoretical analysis and extensive experimental results demonstrate that CP-FL can effectively defend against gradient leakage attacks while maintaining practical utility.

Since the hybrid beamforming (HBF) can approach the performance of fully-digital beamforming (FDBF) with much lower hardware complexity, we investigate the HBF design for beam-hopping (BH) low earth orbit (LEO) satellite communications (SatComs). Aiming at maximizing the sum-rate of totally illuminated beam positions during the whole BH period, we consider joint beamforming and illumination pattern (BIP) design subject to the HBF constraints and sum-rate requirements. To address the nonconvexity of the HBF constraints, we temporarily replace the HBF constraints with the FDBF constraints. Then a joint FDBF and illumination pattern design scheme is proposed using random search and fractional programming (FP) methods. Based on the designed illumination patterns, we optimize the digital beamformers with the constrained analog beamformers by utilizing the FP methods, where a sum-rate maximization HBF scheme is proposed. Simulation results show that the proposed schemes can achieve satisfactory sum-rate performance for BH LEO SatComs.

Machine unlearning allows data owners to erase certain data and its impact from learning models for the right to be forgotten. However, privacy risks during the unlearning process have been identified. Earlier studies have used differences in model outputs before and after unlearning to conduct membership inference attacks. Nevertheless, the current attacks on machine unlearning are limited to inference and cannot reconstruct data without access to the victim's dataset. In this paper, we propose a reconstruction attack towards machine unlearning (RAU), which can reconstruct the unlearned data by exploiting the privacy leakage from the two models. To improve reconstruction quality, we propose a Conditional Matching Generative Adversarial Network (CMGAN), a novel variant of generative adversarial networks which introduces a reconstructive loss. Our work demonstrates the possible privacy leakage of current machine unlearning scenarios. Experimental results on MNIST and Fashion-MNIST show that the proposed attack achieves high label recovery accuracy and good data recovery performance.

Federated learning (FL) has become a hot research domain due to its privacy protection for model collaboratively training in edge computing systems. However, recent studies indicated that most FL algorithms have desperately suffered from backdoor attacks. Although many backdoor defence FL algorithms were proposed, their effects were highly related to the ratio of malicious clients (RMC) of all participated edge nodes. To be more specific, most of them only set RMC around 10% to 30% in their experiments, and their results also showed that the rate of successful backdoor defence seriously drops when RMC increases. In the paper, we propose a novel federated learning scheme with mode connectivity (FedMC) to defend against backdoor attacks, mitigating the sharp defence effect degradation as RMC increases. Conventional mode connectivity mainly focuses on training a connecting curve between two end models, which is inapplicable in distributed multiple clients FL situations. We extend the two-ends mode connectivity to multi-ends by introducing a scalable regularization term consisting of the edge clients' models to involve their knowledge in the connective model training. In each communication round, the FL-Server aggregates and absorbs the contribution of clients by training a connective model based on a small set of clean samples, which builds a pathway to accurately connect all edge clients' models and mitigates the backdoor triggers of models. Extensive experiments and results demonstrate that FedMC can effectively defend against backdoor attacks while maintaining the accuracy on untampered test data.

In vehicular networks, the Point of Interest (POI) query was widely used in Location-Based Services (LBS) for vehicle applications. However, since the attackers can easily access the location, query content, and other information, it is difficult to protect the LBS privacy of vehicle users only using location privacy protection or query privacy protection independently. Therefore, we propose a location privacy and query privacy joint protection scheme based on dummy sequences. According to the limitations of the POI query, we model the correlations between location privacy and query privacy into semantic correlation, temporal correlation, and spatial-temporal attribute correlation, characterized by the Euclidean distance and the association rule algorithm. Moreover, we formulate the dummy sequence selection as a constrained multi-objective optimization problem to obtain the query sequence with a high level of anonymity and a big cloaking region. Experimental results demonstrate that our scheme can resist joint attacks on location privacy and query privacy and protect users' LBS privacy more efficiently compared to the existing schemes.

Graphs provide a unique representation of real- world data. However, recent studies found that inference attacks can extract private property information of graph data from trained graph neural networks (GNNs), which arouses privacy concerns about graph data, especially in collaborative learning systems where model information is more accessible. While there has been a few research efforts on the property inference attacks against GNNs, how to defend against such attacks has seldom been studied. In this paper, we propose to leverage the information bottleneck (IB) principle to defend against the property inference attacks. Particularly, we involve a threat model, where the attacker can extract graph property from the graph embedding developed by GNNs. To defend against the attacks, we use IB to construct new graph structures from the original graphs. The change in graph structures enables the new graphs to contain less information related to the property information of the original graphs, making it harder for attackers to infer property information of the original graphs from the graph embeddings. Meantime, the IB principle enables task-relevant information to be sufficiently contained in the new graph, enabling GNNs to develop accurate predictions. The experimental results demonstrate the efficacy of the proposed approach in resisting property inference attacks and developing accurate predictions.

Machine unlearning has quickly emerged as a technique to withdraw users’ data from the trained model to protect their privacy. Yet the cost of completely unlearning by retraining is high and the unlearning service is thus hard to proceed in the market. We work on the SISA method that greatly lowers the cost of unlearning as an example in this manuscript. We model the problem with a game theory model that balances customers’ benefit and the service provider’s profit, and the optimal price is acquired that both parties could accept. More specifically, in the game model, we calculate customers’ average waiting time with bulk service queueing model, and linearly estimate the customers’ benefit in terms of privacy from withdrawing their data. Our method RUE shows that both parties get more benefit or profit than others, which could make the unlearning service run smoothly when the concerns about the high price are removed. We also analyse the influence of the waiting time on the number of unlearning requests and on the price.

Abstract Publish/subscribe(pub/sub) systems are widely used in large-scale messaging systems due to their asynchronous and decoupled nature. With the population of pub/sub cloud services, the privacy protection problem of pub/sub systems has started to emerge, and events and subscriptions are exposed when executing event matching on untrustworthy cloud brokers. However, as the number of subscriptions increases, the effectiveness of the previous confidentiality protection approaches declines drastically. In this paper, we propose SBM (scalable blind matching), an effective confidentiality protection scheme for pub/sub systems. To the best of our knowledge, SBM is the first scheme that applies order-preserving encryption algorithm to protect the system’s confidentiality and ensure its scalability. In this scheme, SBM-I is highly effective in subscription matching but is unable to achieve ideal security IND-OCPA, whereas SBM-II is suggested to ensure system security and SGX is used to reduce interaction and boost ciphertext matching performance. The experiment demonstrates that this method has better matching performance compared to others: the average matching time of SBM-I is 3–4 orders of magnitude faster than the matching algorithm MP and SGX-based algorithm SCBR when the number of subscriptions is 500,000, and the average matching time of SBM-II is 40 times faster than MP and 24 times than SCBR.

Internet of Things (IoT) devices can utilize deep learning (DL) to boost their intelligence, but also suffer from the long model training process. IoT devices thus may reuse public pretrained models to expedite the training through transfer learning. However, pretrained models may be subject to model-reuse attacks initiated by malicious DL servers, causing models to misclassify targeted data, which poses a threat to the security of IoT devices. In this work, we propose a new model usability detection scheme, the defense against model-reuse attacks (DMRAs), suitable for IoT scenarios. DMRA employs a variant of Lagrange’s mean value theorem to reverse-check the model, which is computationally efficient, thus, suitable for resource-constrained devices. Experimental evaluations on different data sets first validate that model-reuse attacks can attack models in federated learning. And, then demonstrate that DMRA detects such insidious attacks with up to 80% success rate at a lightweight computational cost.

There are increasing concerns about malicious attacks on autonomous vehicles. In particular, inaudible voice command attacks pose a significant threat as voice commands become available in autonomous driving systems. How to empirically defend against these inaudible attacks remains an open question. Previous research investigates utilizing deep learning-based multimodal fusion for defense, without considering the model uncertainty in trustworthiness. As deep learning has been applied to increasingly sensitive tasks, uncertainty measurement is crucial in helping improve model robustness, especially in mission-critical scenarios. In this article, we propose the multimodal fusion framework (MFF) as an intelligent security system to defend against inaudible voice command attacks. MFF fuses heterogeneous audio–vision modalities using VGG family neural networks and achieves the detection accuracy of 92.25% in the comparative fusion method empirical study. Additionally, extensive experiments on audio–vision tasks reveal the model’s uncertainty. Using expected calibration errors, we measure calibration errors and Monte Carlo Dropout to estimate the predictive distribution for the proposed models. Our findings show empirically to train robust multimodal models, improve standard accuracy and provide a further step toward interpretability. Finally, we discuss the pros and cons of our approach and its applicability for advanced driver assistance systems.

As a promising distributed machine learning paradigm, federated learning (FL) trains a central model with decentralized data without compromising user privacy, which makes it widely used by Artificial Intelligence Internet of Things (AIoT) applications. However, the traditional FL suffers from model inaccuracy, since it trains local models only using hard labels of data while useful information of incorrect predictions with small probabilities is ignored. Although various solutions try to tackle the bottleneck of the traditional FL, most of them introduce significant communication overhead, making the deployment of large-scale AIoT devices a great challenge. To address the above problem, this article presents a novel distillation-based FL (DFL) method that enables efficient and accurate FL for AIoT applications. By using knowledge distillation (KD), in each round of FL training, our approach uploads both the soft targets and local model gradients to the cloud server for aggregation, where the aggregation results are then dispatched to AIoT devices for the next round of local training. During the DFL local training, in addition to hard labels, the model predictions approximate soft targets, which can improve model accuracy by leveraging the knowledge of soft targets. To further improve our DFL model performance, we design a dynamic adjustment strategy of loss function weights for tuning the ratio of KD and FL, which can maximize the synergy between soft targets and hard labels. Comprehensive experimental results on well-known benchmarks show that our approach can significantly improve the model accuracy of FL without introducing significant communication overhead.

Although the next generation networks (5G-NGNs) provide a flexible infrastructure to support latency-sensitive and bandwidth-hungry mission-critical Internet of Things (IoT) applications, however, the 5G-IoT integration in NGNs has increased the threat surface. Unfortunately, IoT devices are resource constrained, and the traditional intrusion detection systems (IDS) approaches based on cryptography are not effective on 5G-IoT ecosystems. In this article, we propose an effective 5G-IoT node authentication approach that leverages unique radio frequency (RF) fingerprinting data to train the Deep learning model to detect legitimate and nonlegitimate IoT nodes. Our approach is based on Mahalanobis Distance theory and Chi-square distribution theories. The proposed approach achieves a higher detection accuracy (99.35%) as well as lower training time compared to other existing approaches which is a key benefit of our approach in NGNs. The experiments are conducted using ETSI-open source NFV management and orchestration (OSM-MANO) platform on Amazon Web Services (AWSs) cloud platform to verify how the proposed approach would fit in real-life scenarios. The method can be used as a standalone security system or as a part of multifactor authentication.

Massive Internet of Things (IoT) devices have been playing a critical role in both the cyber and physical worlds. Various cyber attacks pose significant risks to IoT. Machine learning-based intrusion detection system (IDS) has earned much research attention. However, the intrusion prevention system (IPS) is rarely explored. Realtime intrusion prevention is quite challenging because the decision has to be made during a flow rather than after it finishes. Restricted by aligning with the shortest flows, existing IPSs generally inspect only the very first packets, leading to information loss for accurate detection. In this article, we first measure the information loss quantitatively. Then we devise Sniper, an IoT IPS scheme consisting of a flow length predictor, a novel feature space, and an enhanced ensemble learning algorithm. The flow length predictor guides a proper prevention time point to preserve as much information as possible. The proposed Markov matrix-based feature encoding method further saves more information than existing ones. The enhanced learning algorithm ensures a low-false positive rate (FPR), which is critical for IPSs. We benchmark Sniper with one closed-world and three open-world data sets. The results show that Sniper achieves a 99.89% prevention rate and 0.03% FPR, which is superior to the five state-of-the-art baseline models.

The vision of heterogeneous vehicle networks (HetVNETs) embraces various highly dynamic scenarios with urgent requirements for delay-sensitive and reliability-guaranteed computation placement. Incorporating mobile-edge computing (MEC) technology into computation placement has a significant potential to reduce computational delay and enhance communication reliability. However, vehicle mobility and resource constraints make the multivehicle scramble for communication and computational resources challenging. This article intends to investigate collaborative computing by comprehensively considering vehicle mobility, channel condition, and computational resources with two goals: 1) high-reliability transmission (HRT) and 2) computational delay minimization (CDM). Specifically, we develop a hybrid MEC-enabled computation placement orchestrator for HetVNET, where the HRT and CDM are formulated as mixed-integer programming and nonconvex optimization problems, respectively. To ensure high-reliability communication, we leverage the conditional value at risk theory to tackle the nonsmooth HRT problem. To solve the CDM problem, we transform it into two subproblems: resource allocation and task offloading problems, aiming at reducing computational delay and improving resource utilization. Furthermore, we construct an iterative optimization algorithm to capture the optimal computation placement scheme in closed form for the HRT and CDM problems. Performance evaluations show that the proposed methods can significantly improve communication reliability and reduce computational delay.

As a result of the rapid development of Internet of Things (IoT) systems, an increasing number of academics are focusing on finding new applications for IoT systems. For IoT systems, crowdsourcing is a prevalent practise. Due to the large number of deployed devices in IoT networks, more research is still required on the privacy and trust issues that arise when utilizing crowdsourcing. As a result of the characteristics of social computing, the crowdsourcing network poses issues in terms of confidentiality and reliability. To consolidate and create this industry, we have built a differentially private crowdsourcing system that integrates public and private blockchains to address the privacy and trust issues of conventional crowdsourcing systems. Our proposed solution enables varying levels of privacy protection to protect the user’s identity and location. Moreover, the installation of blockchain networks might potentially ensure the data’s integrity. In the conclusion of this article, the possibility of deploying a crowdsourcing system with blockchain in IoE networks is examined.

In the scenario of Industry 4.0, mobile smart devices (SDs) on production lines have to process massive amounts of data. These computing tasks sometimes far exceed the computing capability of SDs and require lots of energy and time to process. How to effectively reduce energy consumption and latency is necessary to be solved. To this end, we first propose a software-defined network (SDN)-based mobile edge computing (MEC) system. In the MEC system, SDs can offload computation tasks to edge servers to decrease the processing latency and avoid the waste of energy. At the same time, taking advantage of SDN’s programmability, scalability, and isolation of the control plane and the data plane, an SDN controller can manage edge devices within the MEC system. Second, based on a stochastic game, we study the computation offloading and resource allocation problems in the MEC system and establish a stochastic game-based computation offloading model. Furthermore, we prove that the multiuser stochastic game in this system can achieve Nash Equilibrium. We further consider each SD as an independent agent and design a stochastic game-based resource allocation algorithm with prioritized experience replays (SGRA-PERs) to minimize energy consumption and processing latency with Multiagent Reinforcement Learning. Experiment results demonstrate that the proposed SGRA-PER is superior to MADDPG,$Q$-Mix, and MAPPO algorithms, which can significantly reduce the processing delay and energy consumption with dynamic resource allocation. Moreover, SGRA-PER can still keep a higher performance under the increase of SDs, which can be applied in a large-scale MEC system.

As trajectory plays an important role in intelligent transportation systems, achieving trajectory recovery from gradients is of great value. Existing research has shown that federated learning is vulnerable to attacks that recover the original training data from shared gradients. Still, we find that gradients attack is difficult to succeed with trajectory data. Most existing attacks have minimal effectiveness when facing models with higher nonlinearity and temporal-related characteristics, such as destination prediction models. In this article, we propose deep leakage from gradients with mobility prior (DLGMP) algorithm to solve these problems. The proposed DLGMP algorithm leverages the spatiotemporal structural information as the prior mobility knowledge, narrowing down the initial search space sharply and decreasing the difficulty of recovery. We further improve our algorithm with an easily extensible regularization term and an adversarial loss of Wasserstein GAN (WGAN) to help recover accurate and reasonable trajectories. Experiments on three mainstream data sets show good performance of our DLGMP algorithm. We also discuss several possible solutions to prevent gradients attack.

Network partitioning is recognized as an effective auxiliary approach for solving transportation tasks on large-scale traffic networks in a domain-decomposition (DD) manner. Most of the existing related partitioning algorithms are explicitly designed to traffic management problems and merely focus on the implied topology of the networks. In this article, toward the practical problems that happened to traffic forecasting (TF) tasks, we propose a network-partitioning-based DD framework to improve graph convolutional network (GCN)-based predictors’ performance on large-scale transportation networks. Particularly, we devise a data-driven network-partitioning approach, namely, speed-matching-partitioning (SMP), which employs not only the topological features but also the traffic speed observations of traffic networks for partitioning. Additionally, we propose a data-parallel training strategy that feeds partitioned subnetworks into independent predictors for parallel training. The proposed approach is tested by comprehensive case studies on three real-world data sets to evaluate its effectiveness. The results indicate that the proposed approach can help improve GCN-based predictors’ accuracy and training efficiency on both small and relatively large traffic data sets. Furthermore, we investigate the model sensitivity to the selection of graph representations and framework parameters, and the learning efficiency of the data-parallel training strategy.

Deep learning has achieved significant breakthroughs in speaker recognition, driven by continual advancements in foundation models. However, malicious third-party platforms have introduced a severe security concern through backdoor attacks, in which attackers can manipulate a model to output a specific label by implanting a trigger. Existing speech backdoor attack methods typically utilize fixed and unnoticeable perturbations as triggers, but these may still be audible and thus detected during training and inference stages. To overcome this limitation, we propose a novel backdoor attack paradigm (PhaseBack) injecting triggers in the phase spectrum. PhaseBack exhibits sufficient stealth by leveraging the fact that the human ear is insensitive to phase information. Besides, injecting partial perturbations in the frequency domain results in global perturbations throughout the time domain, making the attack more effective. Extensive experiments on the Voxceleb1 dataset demonstrate the effectiveness and stealthiness of PhaseBack. Moreover, it has strong resistance to bypass several defense methods.

BCube is a promising structure of data center network, as it can significantly improve the performance of typical applications. With the expansion of network scale and increasement of complexity, reliability and stability of networks have become more essential. In this paper, we study the fault-tolerant routings in BCube. First, we design a fault-tolerant routing algorithm based on node disjoint multi-paths. The proposed multi-path routing has stronger fault tolerance, since each path has no other common nodes except the source node and the destination node. Second, we investigate an effective fault-tolerant routing based on routing capabilities algorithm for BCube. The proposed algorithm has higher fault tolerance and success rate of finding feasible routes, since it does not limit the faults number. Third, we present an adaptive path finding algorithm for establishing virtual links between any two nodes in BCube, which can shorten the diameter of BCube. Extensive simulation results show that the proposed routing scheme outperforms the existing popular algorithms. Compared with the state-of-the-art fault-tolerant routing algorithms, the proposed algorithm has a 21.5% to 25.3% improvement on both throughput and packet arrival rate. Meanwhile, it reduces the average latency of 18.6% and the maximum latency of 23.7% in networks.

Geo-Distributed Machine Learning (Geo-DML) has been proposed to collaborate geographically dispersed data centers (DCs) and train large scale machine learning (ML) models for various applications. While Geo-DML can achieve excellent performance, it also injects massive data traffic into the Wide Area Networks (WANs) in order to exchange gradients during model training process. Such a huge amount of traffic will not only incur network congestion and prolong the training procedure, but also result in straggler problem when DCs are working in heterogeneous network environments. To alleviate these problems, we propose Self-Adaptive Gradient Quantization (SAGQ) for Geo-DML in this work. In SAGQ, each worker DC adopts specific quantization method based on the heterogeneous and dynamic link bandwidth in order to reduce the communication overhead and balance the communication time among worker DCs. By doing so, SAGQ will speed up the Geo-DML training process without sacrificing the ML model performance. Extensive experiments show that compared with the state-of-the-art techniques, SAGQ reduces the Wall-clock time spent to train an ML model by 1.13×–21.31×. In addition, SAGQ can also improve the model accuracy by 0.11%–2.27% over baselines.

In group decision making (GDM), opinion dynamics is a useful tool to investigate consensus formation. Notably, consensus convergence speed is of key importance to manage the consensus formation in GDM with opinion dynamics. Recently, social network DeGroot (SNDG) model has been widely used in opinion dynamics. Based on this, this article dedicates to study how agents’ high self-confidence levels affect the consensus convergence speed in SNDG model. Interestingly, using theoretical analysis, we prove that: 1) the speed of consensus reaching is subject to the largest self-confidence level of opinion followers and 2) the speed of consensus reaching is also subject to the top two self-confidence levels of opinion leaders. Furthermore, through extensive simulation’, we find that the theoretical results are robust to the topological structure and the size of social networks.

Despite achieving exceptional performance, deep neural networks (DNNs) suffer from the harassment caused by adversarial examples, which are produced by corrupting clean examples with tiny perturbations. Many powerful defense methods have been presented such as training data augmentation and input reconstruction which, however, usually rely on the prior knowledge of the targeted models or attacks. A clean example and its adversarial version are very similar but have different high-level representations in a victim model. If we can obtain a space in which the representations of similar examples are also similar, then adversarial examples can be picked out by comparing the representations of input examples in this space and the high-level space of the victim model. Inspired by this, we propose a novel approach for detecting adversarial images, which can protect any pre-trained DNN classifiers and resist an endless stream of new attacks. Specifically, we first adopt a dual autoencoder to project images to a latent space. The dual autoencoder uses the self-supervised learning to ensure that small modifications to samples do not significantly alter their latent representations. Next, the mutual information neural estimation is utilized to enhance the discrimination of the latent representations. We then leverage the prior distribution matching to regularize the latent representations. To easily compare the representations of examples in the two spaces, and not rely on the prior knowledge of the targeted model, a simple fully connected neural network is used to embed the learned representations into an eigenspace, which is consistent with the output eigenspace of the targeted model. Through the distribution similarity of an input example in the two eigenspaces, we can judge whether the input example is adversarial or not. Extensive experiments on MNIST, CIFAR-10, and ImageNet show that the proposed method has superior defense performance and transferability than state-of-the-arts.

In cryptocurrencies, privacy of users is preserved using pseudonymity . However, it has been shown that pseudonymity does not result in anonymity if a user's transactions are linkable. This makes cryptocurrencies vulnerable to deanonymization attacks. The current solutions proposed in the literature suffer from at least one of the following issues: (1) requiring a trusted third–party entity, (2) poor performance, and (3) incompatible with the standard structure of cryptocurrencies. In this article, we propose Unlinkable Coin (UCoin), a secure mix–based approach to address these issues. In UCoin, the link between the input (payer) and output (payee) addresses in a transaction is broken. This is done by mixing the transactions of multiple users into a single aggregated transaction in which the output addresses have been secretly shuffled. In our protocol design, we first develop HDC–net, a secure shuffling protocol that enables a group of users to anonymously publish their data. Then, we deploy the proposed HDC–net protocol in the UCoin architecture (as a mixing unit) to generate the aggregate transactions. We show that UCoin (1) does not rely on a trusted third–party, (2) can mix 50 transactions in 6.3 seconds that is 18% faster than the current solutions, and (3) is fully compatible with the architecture of cryptocurrencies.

In smart farming sector, Internet of Things (IoT) based smart sensing systems are vulnerable to failure, malfunction, and malicious attacks. Also, sensors are deployed often in an alien and harsh environment. Here, the conditions are not well supportive which either causes the sensor to fail prematurely or gives unusual and erroneous readings, known as outliers. This effects the smart network's performance and decision-making ability in many ways. Therefore, it is important to accurately detect the IoT sensor behaviour in legitimate, faulty, and compromised or attack scenarios. To distinguish the sensor behaviour in different scenarios we have proposed a feasible approach using spatial correlation theory which is validated using Moran'sIindex tool. We have used Classification and Regression Trees (CART), Random Forest (RF), and Support Vector Machine (SVM) models to test our approach. For real-time anomaly detection we have used an edge computing technology. We have compared the proposed approach, using Forest Fire real dataset, with the three existing recent works. Our results are promising in terms of accurate detection of IoT sensor behaviours in real-time. This will assist the precision farming industry in making better decisions to securely manage IoT field network, increase productivity, and improves operational efficiency.

Federated learning has emerged as a promising paradigm by collaboratively training a global model through sharing local gradients without exposing raw data. However, the shared gradients pose a threat to privacy leakage of local data. The central server may forge the aggregated results. Besides, it is common that resource-constrained devices drop out in federated learning. To solve these problems, the existing solutions consider either only efficiency, or privacy preservation. It is still a challenge to design a verifiable and lightweight secure aggregation with drop-out resilience for large-scale federated learning. In this article, we propose VOSA, an efficient verifiable and oblivious secure aggregation protocol for privacy-preserving federated learning. We exploit aggregator oblivious encryption to efficiently mask users’ local gradients. The central server performs aggregation on the obscured gradients without revealing the privacy of local data. Meanwhile, each user can efficiently verify the correctness of the aggregated results. Moreover, VOSA adopts a dynamic group management mechanism to tolerate users’ dropping out with no impact on their participation in future learning process. Security analysis shows that the VOSA can guarantee the security requirements of privacy-preserving federated learning. The extensive experimental evaluations conducted on real-world datasets demonstrate the practical performance of the proposed VOSA with high efficiency.

Internet of Things (IoT) devices upload their data into the cloud for storage because of their limited resources. However, cloud storage data has been subject to potential integrity threats, and consequently auditing techniques are demanded to ensure the integrity of stored data. Unfortunately, existing auditing approaches require owners to undertake expensive tag calculations, which is unsuitable for resource-constrained IoT devices. To resolve the issue, we present aFairCloudAuditing proposal by employing theBlockchain (FCAB). We combine certificateless signatures with the designed dynamic structure to constructively offload the cost of tag computation from the IoT device to the introduced fog node, significantly reducing the local burden. Considering that fog nodes may behave dishonestly during auditing, FCAB enables the IoT device to verify the audit result's authenticity by extracting reliable checking records from the blockchain, thereby achieving auditing fairness, which ensures that thehonestcloud and fog node will gain the corresponding reward. Finally, FCAB is proved to satisfy tag unforgeability, proof unforgeability, privacy preserving, and auditing fairness. Experiment evaluations affirm that FCAB is computationally and communicationally efficient and retains a smaller and fixed computation locally at the data processing stage (mainly including tag computation) than existing auditing methods.

Federated learning (FL) empowers distributed clients to collaboratively train a shared machine learning model through exchanging parameter information. Despite the fact that FL can protect clients’ raw data, malicious users can still crack original data with disclosed parameters. To amend this flaw, differential privacy (DP) is incorporated into FL clients to disturb original parameters, which however can significantly impair the accuracy of the trained model. In this work, we study an imperative question which has been vastly overlooked by existing works: what are the optimal numbers of queries and replies in FL with DP so that the final model accuracy is maximized. In FL, the parameter server (PS) needs to query participating clients for multiple global iterations to complete training. Each client responds a query from the PS by conducting a local iteration. We consider FL that will uniformly and randomly select participating clients to conduct local iterations with the FedSGD algorithm. Our work investigates how many times the PS should query clients and how many times each client should reply the PS by incorporating two most extensively used DP mechanisms (i.e., the Laplace mechanism and Gaussian mechanisms). Through conducting convergence rate analysis, we can determine the optimal numbers of queries and replies in FL with DP so that the final model accuracy can be maximized. Finally, extensive experiments are conducted with publicly available datasets: MNIST and FEMNIST, to verify our analysis and the results demonstrate that properly setting the numbers of queries and replies can significantly improve the final model accuracy in FL with DP.

Albeit the popularity of federated learning (FL), recently emerging model-inversion and poisoning attacks arouse extensive concerns towards privacy or model integrity, which catalyzes the developments of secure federated learning (SFL) methods. Nonetheless, the collisions between its privacy and integrity, two equally crucial elements in collaborative learning scenarios, are relatively underexplored. Individuals’ wish to “hide in the crowd” for privacy frequently clashes with aggregators’ need to resist abnormal participants for integrity (i.e., the incompatibility between Byzantine robustness and differential privacy). The dilemma prompts researchers to reflect on how to build mutual confidence between individuals and aggregators. Against the backdrop, this paper proposes a multi-shuffler secure federated learning (MSFL) framework, based on which we further propound three modules (hierarchical shuffling mechanism, malice evaluation module, and composite defense strategy) to jointly guarantee strong privacy protection, efficient poisoning resistance, and agile adversary elimination. Extensive experiments on standard datasets exhibited the method's effectiveness in thwarting different FL poisoning attack paradigms with a minimal cost of privacy breaches.

Insider threat is destructive and concealable, making addressing it a challenging task in cybersecurity. Most existing methods transform user behavior into sequential information and analyze user behavior while neglecting structural information among users, resulting in high false positives. To solve this problem, in this paper, we propose Dual-Domain Graph Convolutional Network (referred to as DD-GCN), a graph-based modularized method for high accuracy and adaptive insider threat detection. The central idea is to convert user features and structural information into heterogeneous graphs in the light of various relationships and take user behavior and relationship into account together. To this end, a weighted feature similarity mechanism is applied to balance the feature similarity of users and original linkages among them so as to generate the fused structure. Next, specific graph embeddings are extracted from the original topology structure and fused structure simultaneously, which convert behavior information into high-level representations. Furthermore, an attention mechanism is applied to learn the adaptive importance weights of the user’s features in the corresponding embedding. The combination and difference constraints are proposed to enhance the learned embeddings’ commonality and the ability to capture different information. Extensive experiments on two real-world datasets clearly show that our proposed DD-GCN extracts the most correlated information from structural topology and feature information substantially, and achieves improved accuracy with a clear margin.

Industrial Internet of Things (IIoT), which has the capability of perception, monitoring, communication and decision–making, has already exposed more security problems that are easy to be invaded by malware because of many simple edge devices that help smart factories, smart cities and smart homes. In this paper, a two–layer malware spread–patch modelIIPVis proposed based on a hybrid patches distribution method according to the simple edge equipments and limited central computer resources of IIoT. The spread process of malware in IIoT was deeply analyzed using differential game and a differential game model was established. Then optimization theory was further used to solve the optimization problem extracted by introducing subjective effort parameters to obtain the optimal control strategies of devices for malware and patches. In addition, we combined the deep reinforcement learning algorithm into the modelIIPVto design a new algorithmDDQN–PVsuitable for suppressing the spread of malware in IIoT during the experiments. Finally, the effectiveness of modelIIPVand algorithmDDQN–PVare verified by numerous comparative experiments.

Advanced persistent threat (APT) is a kind of stealthy, sophisticated, and long-term cyberattack that has brought severe financial losses and critical infrastructure damages. Existing works mainly focus on APT defense under stable network topologies, while the problem under time-varying dynamic networks (e.g., vehicular networks) remains unexplored, which motivates our work. Besides, the spatiotemporal dynamics in defense resources, complex attackers’ lateral movement behaviors, and lack of timely defense make APT defense a challenging issue under time-varying networks. In this paper, we propose a novel game-theoretical APT defense approach to promote real-time and optimal defense strategy-making under both periodic time-varying and general time-varying environments. Specifically, we first model the interactions between attackers and defenders in an APT process as a dynamic APT repair game, and then formulate the APT damage minimization problem as the precise prevention and control (PPAC) problem. To derive the optimal defense strategy under both latency and defense resource constraints, we further devise an online optimal control-based mechanism integrated with two backtracking-forward algorithms to fastly derive the near-optimal solution of the PPAC problem in real time. Extensive experiments are carried out, and the results demonstrate that our proposed scheme can efficiently obtain optimal defense strategy in 54481 ms under seven attack-defense interactions with 9.64% resource occupancy in stimulated periodic time-varying and general time-varying networks. Besides, even under static networks, our proposed scheme still outperforms existing representative APT defense approaches in terms of service stability and defense resource utilization.

Nowadays, many edge computing service providers expect to leverage the computational power and data of edge nodes to improve their models without transmitting data. Federated learning facilitates collaborative training of global models among distributed edge nodes without sharing their training data. Unfortunately, existing privacy-preserving federated learning applied to this scenario still faces three challenges: 1) It typically employs complex cryptographic algorithms, which results in excessive training overhead; 2) It cannot guarantee Byzantine robustness while preserving data privacy; and 3) Edge nodes have limited computing power and may drop out frequently. As a result, the privacy-preserving federated learning cannot be effectively applied to edge computing scenarios. Therefore, we propose a lightweight and secure federated learning scheme LSFL, which combines the features of privacy-preserving and Byzantine-Robustness. Specifically, we design the Lightweight Two-Server Secure Aggregation protocol, which utilizes two servers to enable secure Byzantine robustness and model aggregation. This scheme protects data privacy and prevents Byzantine nodes from influencing model aggregation. We implement and evaluate LSFL in a LAN environment, and the experiment results show that LSFL meets fidelity, security, and efficiency design goals, and maintains model accuracy compared to the popular FedAvg scheme.

Empowered by 5G, it has been extensively explored by existing works on the deployment of differentially private federated learning (DPFL) in the Industrial Internet of Things (IIoT). Through federated learning, decentralized IIoT devices can collaboratively train a machine learning model by merely exchanging model gradients with a parameter server (PS) for multiple global iterations. Differentially private (DP) mechanisms will be incorporated by IIoT devices (also called clients) to prevent the leakage of privacy due to the exposure of gradients because original gradients will be distorted DP noises. Yet, learning with distorted gradients can seriously deteriorate model accuracy, making DPFL unusable in reality. To address this problem, we propose a novel DPFL with sparse responses (DPFL-SR) algorithm, which applies the sparse vector technique to reduce the privacy budget consumption in each global iteration. Specifically, DPFL-SR evaluates the value of each gradient, and only distorts and uploads significant gradients to the PS because significant gradients are more essential for model training. Since insignificant gradients are not disclosed, the reserved privacy budget can be used to return significant gradients for more iterations so that DPFL-SR can achieve higher model accuracy without lowering the privacy protection level. Extensive experiments are conducted with the MNIST and Fashion-MNIST datasets to demonstrate the practicability and superiority of DPFL-SR in IIoT systems.

Recent studies indicate that Graph Neural Networks (GNNs) are vulnerable to adversarial attacks. Particularly, adversarially perturbing the graph structure, e.g., flipping edges, can lead to salient degeneration of GNNs’ accuracy. In general, efficiency and stealthiness are two significant metrics to evaluate an attack method in practical use. However, most prevailing graph structure-based attack methods are query intensive, which impacts their practical use. Furthermore, while the stealthiness of perturbations has been discussed in previous studies, the majority of them focus on the attack scenario targeting a single node. To fill the research gap, we present a global attack method against GNNs, Saturation adversarial Attack with Meta-gradient, in this article. We first propose an enhanced meta-learning-based optimization method to obtain useful gradient information concerning graph structural perturbations. Then, leveraging the notion of saturation attack, we devise an effective algorithm to determine the perturbations based on the derived meta-gradients. Meanwhile, to ensure stealthiness, we introduce a similarity constraint to suppress the number of perturbed edges. Thorough experiments demonstrate that our method can effectively depreciate the accuracy of GNNs with a small number of queries. While achieving a higher misclassification rate, we also show that the perturbations developed by our method are not noticeable.

Edge caching has been regarded as a promising technique for low-latency, high-rate data delivery in future networks, and there is an increasing interest to leverage Machine Learning (ML) for better content placement instead of traditional optimization-based methods due to its self-adaptive ability under complex environments. Despite many efforts on ML-based cooperative caching, there are still several key issues that need to be addressed, especially to reduce computation complexity and communication costs under the optimization of cache efficiency. To this end, in this paper, we propose an efficient cooperative caching (FDDL) framework to address the issues in mobile edge networks. Particularly, we propose a DRL-CA algorithm for cache admission, which extracts a boarder set of attributes from massive requests to improve the cache efficiency. Then, we present an lightweight eviction algorithm for fine-grained replacements of unpopular contents. Moreover, we present a Federated Learning-based parameter sharing mechanism to reduce the signaling overheads in collaborations. We implement an emulation system and evaluate the caching performance of the proposed FDDL. Emulation results show that the proposed FDDL can achieve a higher cache hit ratio and traffic offloading rate than several conventional caching policies and DRL-based caching algorithms, and effectively reduce communication costs and training time.

When making decisions, individuals often express their preferences linguistically. The computing with words methodology is a key basis for supporting linguistic decision making, and the words in that methodology may mean different things to different individuals. Thus, in this article, we propose a continual personalized individual semantics learning model to support a consensus-reaching process in large-scale linguistic group decision making. Specifically, we first derive personalized numerical scales from the data of linguistic preference relations. We then perform a clustering ensemble method to divide large-scale group and conduct consensus management. Finally, we present a case study of intelligent route optimization in shared mobility to illustrate the usability of our proposed model. We also demonstrate its effectiveness and feasibility through a comparative analysis.

Planetary gearboxes (PGs) are significant transmission chains in industrial applications. Scientifically perceiving its operational security domain (OSD) is crucial for reliability analysis and preventive maintenance. Unfortunately, most security domain analysis methods discard the abundant information from the fault diagnosis indicator and measured data. To this end, this article proposes a frequency energy ratio (FER) based strategy to intelligently diagnose gear faults, identify the OSD and perform a reliability analysis of the OSD. The FER cell is constructed to consider multiple sideband numbers and bandwidth conditions. The optimal parameter pair among FER cells is determined by a novel distance measurement strategy, namely intra-inter class distance metric, which combines deep autoencoder and relatively max–minimum distance metric. The optimal FER can be conveniently applied to diagnose the remaining data. More importantly, the optimal FER with different fault-induced frequencies is picked to identify the OSD via minor training data. The failure probability of a large number of vibration data is calculated to verify the effectiveness of the OSD. Experimental results demonstrate that FER owns stronger robustness and diagnostic accuracy than the traditional sideband energy ratio. The proposed methods bridge the fault diagnosis and reliability analysis, which have a bright perspective in perceiving operational conditions from research and engineering fields.

Machine learning-based DDoS attack detection methods are mostly implemented at the packet level with expensive computational time costs, and the space cost of those sketch-based detection methods is uncertain. This paper proposes a two-stage DDoS attack detection algorithm combining time series-based multi-dimensional sketch and machine learning technologies. Besides packet numbers, total lengths, and protocols, we construct the time series-based multi-dimensional sketch with limited space cost by storing elephant flow information with the Boyer-Moore voting algorithm and hash index. For the first stage of detection, we adopt CNN to generate sketch-level DDoS attack detection results from the time series-based multi-dimensional sketch. For the sketch with potential DDoS attacks, we use RNN with flow information extracted from the sketch to implement flow-level DDoS attack detection in the second stage. Experimental results show that not only is the detection accuracy of our proposed method much close to that of packet-level DDoS attack detection methods based on machine learning, but also the computational time cost of our method is much smaller with regard to the number of machine learning operations.

Federated learning (FL) is a machine learning framework for collaborative data mining in many scenarios (e.g. Internet of Things) due to its privacy-preserving feature. However, various attacks arise security concerns of FL, such as poisoning, backdoor, and DDoS attacks. Several blockchain-based FL schemes strengthen credibility and security without considering the increased communication overhead. Some existing work compresses local updated gradients to sign vectors to lower communication overhead at the expense of model accuracy. To address the above concerns, this paper offers a blockchain-based asynchronous SignSGD (BASS) scheme. A novel asynchronous sign aggregation algorithm is introduced to ensure model accuracy even if the local updated gradients are compressed to sign vectors. Considering the unstable network connection on IoT, a consensus algorithm that elects multiple leader nodes enables reliable global model aggregation. The introduced blockchain improves credibility and security without downgrading efficiency. Empirical studies show that BASS outperforms other schemes in efficiency, model accuracy, and security.

The Satellite Terrestrial Integrated Network(STIN) has become an indispensable part of the future network. However, the limited resources, long delay communications and highly exposed channels of SGIN are vulnerable to network attacks, which make the access authentication scheme as the first line of defense in network security. Most of the existing access authentication schemes are based on discrete logarithm and large integer factorization problems, which cannot resist quantum attacks, and the number of interactions between entities are too large. Therefore, we propose a lightweight and certificateless anonymous access authentication scheme based on lattice to solve this problem. We introduce Number Theory Research Unit (NTRU) scheme into the key generation process, to improve the utilization rate of equipment resources and ensure the legitimacy of the communication entity. The performance evaluation results demonstrate that our scheme only needs twice satellite-ground interactions to complete mutual authentication.

Federated learning (FL) is an emerging solution for machine learning model training in edge/fog computing systems. Unlike traditional systems that collect and train models on clouds, FL allows multiple edge/fog nodes to train a global model collaboratively without revealing their local data to clouds. Compared with traditional systems, it is inherited with better privacy protection ability. Although the basic privacy protection is inherited in FL, the privacy leakage from shard models is still unsolved. Existing solutions attempt to enhance the privacy of shared model parameters by adding differential privacy (DP) noise. However, these solutions all suffer from accuracy loss and convergence problems owing to the injected noise. In this paper, we propose a novel federated learning protocol to solve the above problem. The model trained on a carefully selected sampling subset can achieve the same level privacy protection as DP while preserving the model accuracy. Experimentally, we proved that our protocol achieves better model accuracy in the same privacy guarantee compared with noise injecting DP methods.

Satellite formation flying with functional payloads can provide flexibility and real-time services for time sensitive missions through the cooperation of intra-satellite wired and inter-satellite wireless communication systems. However, the difference between wired and wireless scheduling mechanisms will increase the instability of forwarding delay at satellite. To solve this problem, we propose a wired and wireless converged scheduling scheme. Firstly, we construct the transmission rate and time division multiple access scheduling model to describe the inter-satellite link, and the IEEE 802.1Qbv scheduling model for wired link respectively. Secondly, we analyze the influence of the position of packet in different time slots on the forwarding delay, to model the relationship between the converged scheduling and forwarding delay. Finally, we formulate the converged scheduling as a zero-one programming problem with minimum jitter, and adopt the genetic-tabu search algorithm to solve it. Simulation results demonstrate that the delay is reduced by 20%, the jitter is not higher than$40\mu\mathrm{s}$.

Artificial intelligence (AI) has emerged as the most promising solution expected to overcome the high degree of abstraction of radio signals and achieve accurate automatic modulation classification (AMC). To further improve the classification performance of the AMC model and enhance its interpretability, the network output layer is modeled as a decision space into which the input data is projected. In this paper, we expand the inter-class angle between the classes with the largest confusion rate to increase the decision space. In addition, we extend the perspective to the softmax layer and evaluate the negative impact of the output distribution range on the confidence difference in the AMC problem. We further propose constraining the norm of the input data to the output layer in combination with prior knowledge of the distribution of modulation signal data. Combining the above two aspects, a Maximum Focal Inter-Class Angular Loss with Norm Constraint (MFICAL-NC) scheme is proposed. The experimental results show that the method can guide the model to obtain a better fitting state and a stronger generalization ability.

In the Internet of Medical Things, the intelligent auxiliary decision-making system uses machine learning algorithms to analyze medical data for disease diagnosis, auxiliary intervention, and analysis and early warning. However, in the process of medical data transmission, processing, and storage, a large amount of private information is also at risk of leakage. Therefore, this article proposes a smart classification and decision-making program in the Internet of Medical Things scenario-SEMMI, which can effectively deal with the risk of data leakage in the process of medical data processing. At the same time, it reduces the huge computing and storage pressure caused by encryption and decryption operations in medical institutions. In the scheme, data collection, processing, transmission, storage and calculation are completed by ciphertext. In addition, in view of the relatively weak computing and storage capabilities of sensor nodes, we use chaos theory to construct a stream cipher algorithm to ensure the security of transmission from sensor to user; the homomorphic encryption algorithm is used to ensure the computability of the ciphertext and the security of storage. Through security analysis, it can be concluded that this scheme can resist attacks from adversaries; at the same time, the experimental results show that the scheme has good performance in terms of calculation, storage overhead, accuracy, and so on.

With the growing proliferation of intelligent end devices and data analytics techniques, real momentum towards the development of smart healthcare networks (SHN) has already been evident. Multiple parties in SHNs continuously exchange medical data in order to achieve a precise diagnosis and process optimization. Privacy issue emerges since medical data are susceptible, while the combination of a series of medical data may lead to further privacy leakage. Adversaries launch unceasingly launch poisoning attacks, a dominant attack to maliciously manipulate data, severely impact the authenticity of the data transmitting over the SHNs, leading to misdiagnosing or even physical damage. In this paper, we propose a personalized differential privacy model built upon blockchain, in which the community density is exploited to customize the degree of privacy protection and inject corresponding noise data. Besides using blockchain as the underlying network architecture to defeat poisoning attacks. The proposed model can guarantee the authentication of the differentially private data, traceability of data, and single-point failure avoidance in SHN. Evaluation and extensive results using real-world data sets demonstrate the superiority of the proposed model.

Federated learning (FL) as a privacy-preserving machine learning (ML) algorithm provides an efficient distributed training paradigm. Existing FL frameworks still suffer from privacy leakage hazards such as membership inference attacks. The current popular defense approaches are mainly based on differential privacy (DP) strategy. However, privacy preservation is undertaken with an inevitable loss of model utility in DP. As a result, it performs miserably in practical deployments. To solve this problem, we modify the FL framework through the information bottleneck (IB) method to attain a trade-off between privacy protection and model utility. Firstly, we adapt the training process on client side by applying IB in the local training. It is intended to squeeze out privacy through the bottleneck. Secondly, we further modify the training process on server side. A validation process is used to evaluate whether the IB-based local training is squeezing out privacy. Clients that extrude the right information will occupy an important place in aggregation phase. Extensive experiments on classic datasets demonstrate the superiority of the proposed scheme in terms of privacy preservation and model utility.

Recently, there is a growing interest in Double-Layered Satellite Networks (DLSN) which integrate Medium-Earth-Orbit (MEO) and Low-Earth-Orbit (LEO) satellites for provision of mobile and personal services. However, it is still in the early stage with several challenges unaddressed, and one of the key problems is the inter-layer link allocations between MEO and LEO satellites, as DLSN topology is dynamically changed over the time and satellites are with the limited number of connections onboard. To this end, we propose a corresponding Inter-layer Link Allocation (ILA) scheme in this paper, taking the visible duration between satellites, transmitting power consumed onboard and geographical distributions of user load into account, aiming to maximize the utilization efficiency of DLSN inter-layer links. Then, we formulate it as a constrained multi-objective linear programming problem and evaluate its performance with other three benchmarks. Numerical results have demonstrated that the proposed ILA scheme can decrease the number of ILL handovers and average inter-satellite distance, with load balanced between LEO and MEO satellites.

The existing Federated Learning (FL) systems encounter an enormous communication overhead when employing GNN-based models for traffic forecasting tasks since these models commonly incorporate enormous number of parameters to be transmitted in the FL systems. In this paper, we propose a FL framework, namely, C lustering-based hierarchical and T wo-step- optimized FL (CTFL), to overcome this practical problem. CTFL employs a divide-and-conquer strategy, clustering clients based on the closeness of their local model parameters. Furthermore, we incorporate the particle swarm optimization algorithm in CTFL, which employs a two-step strategy for optimizing local models. This technique enables the central server to upload only one representative local model update from each cluster, thus reducing the communication overhead associated with model update transmission in the FL. Comprehensive case studies on two real-world datasets and two state-of-the-art GNN-based models demonstrate the proposed framework’s outstanding training efficiency and prediction accuracy, and the hyperparameter sensitivity of CTFL is also investigated.

The widespread use of mobile devices propels the development of new-fashioned video applications like 3D (3-Dimensional) stereo video and mobile cloud game via web or App, exerting more pressure on current mobile access network. To address this challenge, we adopt the crowdsourcing paradigm to offer some incentive for guiding the movement of recruited crowdsourcing users and facilitate the optimization of the movement control decision. In this paper, based on a practical 4G (4th-Generation) network throughput measurement study, we formulate the movement control decision as a cost-constrained user recruitment optimization problem. Considering the intractable complexity of this problem, we focus first on a single crowdsourcing user case and propose a pseudo-polynomial time complexity optimal solution. Then, we apply this solution to solve the more general problem of multiple users and propose a graph-partition-based algorithm. Extensive experiments show that our solutions can improve the efficiency of real-time D2D communication for mobile videos.

The blockchain-enabled smart logistics market is expected to grow worth USD 1620 billion and at a compound annual growth rate of 62.4%. Smart logistics ensures intelligence infrastructure, logistics automation, real-time analysis of supply chain data synchronization of the logistics process, cost transparency, unbroken shipment tracking all the way down to the transportation route, etc. In the smart logistics domain, significant advancement and growth of the Internet of Things (IoT) sensors are evident. However, the connectivity of IoT systems, including Tactile Internet, without proper safeguards creates vulnerabilities that can still be deliberately or inadvertently cause disruption. In view of this, we primarily notice two key issues. Firstly, the logistics domain can be compromised by a variety of natural or man-made activities, which eventually affect the overall network security. Secondly, there are thousands of entities in the supply chain network that use extensive machine-learning algorithms in many scenarios, and they require high-power computational resources. From these two challenges, we note that the first concern can be addressed by adding blockchain to IoT logistic networks. The second issue can be addressed using 6G. This will support 1-μs latency communications, support seamless computing at the edges of networks, and autonomously predict the best optimal location for edge computing. Motivated by this, we have highlighted motivational examples to show the necessity to integrate 6G and blockchain in smart logistic networks. Then, we have proposed a 6G and blockchain-enabled smart logistic high-level framework. We have presented the key intrinsic issues of this framework mainly from the security and resource management context. In this paper, recent state-of-the-art advances in blockchain enabled next-generation smart logistic networks are analyzed. We have also examined why 6G and not 5G would be compatible with the smart network. We have introduced five different use cases of blockchain technology in smart logistics. Later, this paper discusses some important concerns that blockchain in smart logistics might face. We have also provided potential solutions to tackle these concerns.

Summary Community detection is a fundamental research problem in social networks. However, most existing research focuses on homogeneous networks while ignoring the multitopology and attributes in social media. In this article, we propose community detection algorithms based on community kernels to detect high‐quality communities in heterogeneous social networks. It is noticed that the social community has multiple topology structures, as nodes or users in social media networks have multiple attributions. For example, users can be friends and coworkers in a research group simultaneously. Hence, we propose a multilayer and attribute combined measure (MACM), a novel measurement based on the multilayer structure and common neighboring attributes, which includes the similarity measure between nodes and the importance measure for individual node in multilayer networks. Two improved community kernel detection algorithms based on MACM are subsequently proposed. They are the MA‐Greedy, which is based on the greedy algorithm, and the MA‐WeBA, which is a weighted balanced algorithm. The multilayer structure and attributes are comprehensively considered when calculating the similarity and importance of nodes in these strategies. Extensive experimental results on two public data sets demonstrate that the multilayer structure and attribute information can be used to enhance the precision of community detection.

Summary In the age of the Internet of Things (IoT), large numbers of sensors and edge devices are deployed in various application scenarios; Therefore, collaborative learning is widely used in IoT to implement crowd intelligence by inviting multiple participants to complete a training task. As a collaborative learning framework, federated learning is designed to preserve user data privacy, where participants jointly train a global model without uploading their private training data to a third party server. Nevertheless, federated learning is under the threat of poisoning attacks, where adversaries can upload malicious model updates to contaminate the global model. To detect and mitigate poisoning attacks in federated learning, we propose a poisoning defense mechanism, which uses generative adversarial networks to generate auditing data in the training procedure and removes adversaries by auditing their model accuracy. Experiments conducted on two well‐known datasets, MNIST and Fashion‐MNIST, suggest that federated learning is vulnerable to the poisoning attack, and the proposed defense method can detect and mitigate the poisoning attack.

Abstract Deep neural networks are widely adopted powerful tools for perceptual tasks. However, recent research indicated that they are easily fooled by adversarial examples, which are produced by adding imperceptible adversarial perturbations to clean examples. Here the steganalysis rich model (SRM) is utilized to generate noise feature maps, and they are combined with RGB images to discover the difference between adversarial examples and clean examples. In particular, a two‐stream pseudo‐siamese network that fuses the subtle difference in RGB images with the noise inconsistency in noise features is proposed. The proposed method has strong detection capability and transferability, and can be combined with any model without modifying its architecture or training procedure. The extensive empirical experiments show that, compared with the state‐of‐the‐art detection methods, the proposed approach achieves excellent performance in distinguishing adversarial samples generated by popular attack methods on different real datasets. Moreover, this method has good generalization, it trained by a specific adversary can defend against other adversaries effectively.

A large number of Internet of Things (IoT) devices have been interconnected for information collection and exchange. The data are only meaningful if it is captured at the expected location (i.e., the IoT devices or sensors are not removed accidentally or intentionally). This article presents a new algorithm, which cooperatively locates multiple IoT devices deployed in a 3-D space based on pairwise Euclidean distance measurements. When the distance measurement noises are negligible, a new feasibility problem of rank-3 variables is formulated. We solve the problem using the difference-of-convex (DC) programming to preserve the rank-3 constraints, rather than relaxing the constraints, using semidefinite relaxation (SDR). When the distance measurements are corrupted by additive noises and nonlight-of-sight (NLOS) propagation, a maximum-likelihood estimation (MLE) problem is formulated and transformed to a DC program solved with the rank-3 constraints preserved. Simulation results indicate that the proposed approach can achieve satisfactory accuracy results with a low complexity and strong robustness to the irregular topology, poor connectivity, and measurement errors, as compared to existing SDR-based alternatives.

In the era of Internet of Things (IoT), millions of heterogeneous IoT devices generate an explosion of data and services waiting to be discovered. The convergence of IoT with social networks (SIoT) interconnects multiple IoT applications and alleviates the common data sparsity and cold start problems in traditional recommendation systems. However, the social trust relationships may also be very sparse, which affects the accuracy of trust-based recommendation systems. Meanwhile, mobile devices have limited resources and are more vulnerable to malicious attacks in the IoT environment. In order to complete the trust relationship and further improve the trust-based recommendation performance, we propose a reliable and lightweight trust inference model for service recommendation in SIoT, calledTIRec. First, we obtain a comprehensive weighted centrality metric (LGWC) considering both local and global contexts. Based on this, we propose a corresponding lightweight trust path selection algorithm. Then, we present a reliable trust inference calculation algorithm consist of trust propagation and aggregation strategy, which can efficiently resist two common malicious attacks. Finally, we incorporate the rating, direct trust, and indirect trust together into the matrix factorization model, and integrate the influence of truster and trustee to obtain the synthetic model for rating predication. To the best of our knowledge, this article is the first to integrate trust inference algorithm into the trust-based recommendation systems. The extensive experiments are conducted on three real-world data sets, and the results show that ourTIRecmodel performs better than other advanced recommendation models in both “all users” view and “cold start users” view.

With a large number of devices and a wealth of user data sets, the Internet of Things (IoT) has become a great host for federated learning (FL). At the same time, the massive amount of user data in IoT results in desperate demand for privacy preserving. The onion router (Tor) is a promising method to solve the privacy issue in IoT-based FL by user anonymity. However, IoT devices’ resource is too limited to execute the cryptographic operations in Tor. Moreover, network traffics in Tor can be easily controlled by malicious routers with a fake high self-reported bandwidth. In this article, taking advantage of the Tor, we will introduce an anonymous FL framework in IoT called FedTor. To decrease the cryptographic cost in conventional Tor, we propose a lightweight shared key generation scheme for resource-limited IoT devices. Furthermore, we use the difference between the self-reported bandwidth and the bandwidth observed from others to measure the reputation of onion routers. A reputation-based router selection (RBRS) scheme is then brought up to defend traffic control from malicious routers. We conducted extensive simulations to compare FedTor with related works. The results show that the RBRS scheme can decrease the malicious rate of onion routers and the lightweight shared key has a cost advantage over other schemes.

Since Internet of Things (IoT) communications can enjoy many advantages brought by content-centric networking (CCN) in nature, there is an increasing interest on their integration for better information retrieval and distribution. Nevertheless, different from the conventional multimedia traffic of which contents are hardly changed, IoT data are always transient and updated by their producers according to the actual situation. As a result, if without any effective countermeasures, outdated copies are inevitably stored by CCN routers and then distributed to the associated consumers, degrading both caching efficiency and user experience. In fact, most of related policies take little account of information freshness for cached contents, and how to tackle transient IoT data in CCN is still an ignored but crucial issue required for further explorations. Therefore, in this article, we propose an efficient popularity-based cache consistency management scheme, which aims to guarantee freshness of IoT data returned by on-path routers and avoid heavy signalling costs introduced at the same time. Extensive simulations were performed under both real-world scare-free and binary-tree topologies, and corresponding results have proved the efficiency of the proposed scheme in timely evictions of outdated IoT data stored by CCN in-network caching.

Recent advances in computation, communication, and control technologies have revolutionized the way that humans, smart things, and intelligent systems interact and exchange information. Intelligent cyber–physical systems (ICPSs), characterized by the deep complex intertwining process among cyber components for computation and control with intelligent technologies and the dynamic physical components, will fuel this revolution. Examples of ICPS include intelligent automotive and transportation systems, intelligent avionics systems, smart home, smart building, smart community, smart grid, smart healthcare systems, intelligent wearable systems, intelligent energy systems, robotic systems, etc. Bringing machine/deep-learning-based intelligence techniques into CPSs can improve the performance in many aspects. However, this also imposes new security, privacy, and trust challenges, which highlights the need to develop novel methodologies to tackle these challenges. In this special issue, we will publish the following articles.

Facing multidemand tasks and massive heterogeneous resources in an IoT edge computing environment, it is a challenge to obtain reliable and quick response service and allocate application tasks to resource nodes that meet task requirements and user preference. Since IoT edge computing is facing different types of severe attacks, such as message attacks, swing attacks, collusion attack, node attacks, etc., providing a reliable service environment, trust evaluation between edge nodes is necessary. Existing trust computing schemes, however, suffer from a long response period and low malicious detection rate in a dynamic environment. To alleviate these issues, we propose a reliable and efficient task offloading strategy based on the multifeedback trust mechanism (TOSMFTM). First, a reliable and efficient architecture of TOSMFTM is established, which can effectively improve the ability of trust computing and task offloading. Second, according to the broker’s dynamic monitoring of data, a multifeedback trust aggregation model based on time attenuation and interaction frequency is proposed to provide a trusted running environment. Third, a trust weight$k$-means (TWK-means) clustering algorithm is designed based on resource attributes to enhance the reliability of service, and quickly and accurately cluster out resource nodes required by the task. Finally, we construct a task offloading model based on trust clustering to ensure user experience quality and promote system efficiency. Different from existing task processing models, which only focus on task offloading, our method also carries out resource preprocessing, trust evaluation, and resource clustering before task processing. The experiment verifies the effectiveness and feasibility of our TOSMFTM scheme.

To ensure an accurate power allocation against increasing intelligent jamming attacks on the offloading link of computation tasks, we investigate interactions between a cluster head node and an intelligent jammer using a Stackelberg game framework, under the constraint of the total power to use and the limited knowledge of its own channel gain for each player. In this game, the intelligent jammer gathers channel gain information and processes it using a deep neural network (DNN) to infer the accurate jamming power as an attack strategy. The cluster head node also exploits DNN to infer an accurate transmission power as a defense strategy according to the varying channel gain. We model the optimization of the attack and defense strategies using single channel jamming DNN (SJnet), multiple channel jamming DNN (MJnet), single channel sensor DNN (SSnet), and multiple channel sensor DNN (MSnet) for the single (multiple) channel jamming attacks. In addition, we extend the design to the scenario where the intelligent jammer can launch a hybrid mode jamming attack, and propose a DNN Stackelberg game-based defense scheme. Numerical simulation results demonstrate that our proposed mechanism is superior to other power allocation mechanisms under different scenarios in the sensor edge cloud.

With the application of the Internet of Medical Things (IoMT) in minimally invasive surgery (MIS), surgeons now have a better chance at hard-to-treat cases by carrying out more complicated MIS workflows. However, a scheduled surgical workflow is often required to be updated based on the patient’s internal tissue states. Perioperative complications could occur if in-time adjustments are lacking in the operating rooms when needed. To help manage the uncertainty of live surgical workflows in the IoMT environment, we propose a MIS safety improvement framework. It helps surgeons in predicting surgical workflows with limited MIS video frames by embedding our proposed model GuidedNet. To predict future surgical activities, we first build three isomorphic neural networks to capture the spatiotemporal information. Then, we establish a guidance fusion module to handle the contextual information. It guides the GuidedNet to recognize the surgical stage. Moreover, we build a novel joint loss function to train the GuidedNet to predict the future surgical stage. We evaluate the approach on a large data set that contains 80 cholecystectomy videos (Cholec-80) and compare it with the state of the art. Experiments show that the GuidedNet can assist surgeons in carrying out MIS as well as guide the next stage of surgery for improving surgical safety. Comparing to the state of the art, our approach can obtain better predict accuracy (up to 79%) with less computing resource consumption. The result also shows that our approach has a high application prospect in video classification in other Internet of Things scenarios.

Wireless-powered mobile-edge computing (MEC) emerges as a crucial component in the Internet of Things (IoTs). It can cope with the fundamental performance limitations of low-power networks, such as wireless sensor networks or mobile networks. Although computation offloading and resource allocation in MEC have been studied with different optimization objectives, performance optimization in larger-scale systems still needs to be further improved. More importantly, energy efficiency is also a key issue as well as computation offloading and resource allocation for wireless-powered MEC. In this article, we investigate the joint optimization of computation rate and energy consumption under limited resources, and propose an online offloading model to search for the asymptotically optimal offloading and resource allocation strategy. First, the joint optimization problem is modeled as a mixed integer programming (MIP) problem. Second, a deep reinforcement learning (DRL)-based method, energy efficiency computation offloading using probabilistic action generating (ECOPG), is designed to generate the joint optimization policy for computation offloading and resource allocation. Finally, to avoid the curse of dimensionality in large network scales, an action exploration mechanism based on probability is introduced to accelerate the convergence rate by targeted sampling and dynamic experience replay. The experimental results demonstrate that the proposed methods significantly outperform other DRL-based methods in energy consumption, and gain better computation rate and execution efficiency at the same time. With the expansion of the network scale, the improvements become more apparent.

With the fast boom of Internet of Medical Things (IoMT) devices and an increasing focus on personal health, personal health data are extensively collected by IoMT and stored as personal health records (PHRs). PHRs are frequently shared for accurate diagnosis, prognosis prediction, health advice consulting, etc. Since PHRs are highly private, the data-sharing process leads to wide-ranging concerns on privacy leakage and security compromise. Existing research has shown that the centralized systems, as the mainstream mode, are under the great risks. Motivated by this, we propose a consortium blockchain-based PHR management and sharing scheme, which is both security aware and privacy preserving. We adopt the interplanetary file system (IPFS) to store the PHR ciphertext of IoMT. Then, zero-knowledge proof can provide evidence for verifying keyword index authentication on blockchain. Moreover, the scheme jointly leverages modified attribute-based cryptographic primitives and tailor-made smart contracts to achieve secure search, privacy preservation, and personalized access control in IoMT scenarios. Security analysis is conducted to show that the designed protocols attain the expected design goals. This is followed by extensive evaluation results derived from real-world data sets, which demonstrate the superiority of the proposed scheme over current leading ones.

Existing attribute-based encryption (ABE) schemes with revocation to secure the cloud-assisted Internet of Things (IoTs) raise challenges, such as eliminating the need for predefined public parameters in system initialization, performing the encryption and decryption operations efficiently, and achieving adaptive security under standard security assumption. In this article, we address these challenges by proposing an unbounded and efficient revocable ABE scheme with adaptive security for cloud-assisted IoTs. Distinct from the previous approaches in this field, our scheme not only efficiently realizes access control over encrypted data in a fine-grained and revocable way but also is proved to be adaptively secure under standard decision linear assumption. Meanwhile, the parameters do not need to be predefined in the system initialization and thus, our scheme satisfies the unbounded property. Moreover, the monotonic span program (MSP) is elegantly utilized as the access structure to reduce the number of bilinear pairing and exponentiation operations for encryption and decryption. Theoretical performance analysis and experiment evaluation disclose that our proposed scheme owns outstanding feasibility, efficiency, and effectiveness.

Nowadays, Internet of Things (IoT) is an attractive system to provide broad connectivity of a wide range of applications, and clouds are natural promoters. Cloud-assisted IoT combines the advantages of cloud computing and IoT, which is able to collect data from the real world and maximizes the value of the collected data by the means of data sharing and data analysis. Meanwhile, secure and convenient data retrieval in cloud servers becomes an important requirement for both enterprises and individual users. Public-key encryption with search functionality (shorten as PKE-SF) is a widely used cryptographic technique that allows users to retrieve encrypted data without decryption. PKE-SF mainly contains the primitives of public-key encryption with keyword search (PKE-KS), public-key encryption with equality test (PKE-ET), and plaintext-checkable encryption (PCE). In light of the overwhelming variety and multitude of PKE-SF schemes, this survey presents these schemes from different perspectives to provide better comprehension for beginners and advanced researchers. More concretely, this survey concentrates on the state of the art of PKE-SF by analyzing the design rationale, examining the framework and security model, and assessing the existing schemes in accordance with theoretic efficiency, security properties, and experimental performance. Furthermore, we discuss the extensions of traditional PKE-SF schemes which feature with the access control delegation, conjunctive keyword search, certificate-free, and offline keyword guessing attack resilience. Finally, we point out some promising directions for readers.

Federated learning (FL) is widely adopted in traffic forecasting tasks involving large-scale IoT-enabled sensor data since its decentralization nature enables data providers’ privacy to be preserved. When employingstate-of-the-artdeep learning-based traffic predictors in FL systems, the existing FL frameworks confront overlarge communication overhead when transmitting these models’ parameter updates since the modeling depth and breadth renders them incorporating an enormous number of parameters. In this article, we propose a practical FL scheme, namely, Clustering-based hierarchical and Two-step-optimized FL (CTFed), to tackle this issue. The proposed scheme follows adivide et imperastrategy that clusters the clients into multiple groups based on the similarity between their local models’ parameters. We integrate the particle swarm optimization algorithm and devises a two-step approach for local model optimization. This scheme enables only one but representative local model update from each cluster to be uploaded to the central server, thus reduces the communication overhead of the model updates transmission in FL. CTFed is orthogonal to the gradient compression- or sparsification-based approaches so that they can orchestrate to optimize the communication overhead. Extensive case studies on three real-world data sets and threestate-of-the-artmodels demonstrate the outstanding training efficiency, accurate prediction performance, and robustness to unstable network environments of the proposed scheme.

Privacy-preserving transportation mode identification (TMI) is among the key challenges toward future intelligent transportation systems. With recent developments in federated learning (FL), crowdsourcing has emerged as a promising cost-effective data source for training powerful TMI classifiers without compromising users’ data privacy. However, existing TMI approaches have relied heavily on the availability of transportation mode labels, which is often limited in real-world applications. While recent semisupervised studies have partially addressed this issue by assigning pseudolabels to unlabeled data, such practice often degrades classification performance as more unlabeled data are incorporated. In response to this issue, we present a semisupervised FL scheme for TMI termed mean teacher semisupervised FL (MTSSFL). MTSSFL trains a deep neural network ensemble under a novel semisupervised FL framework, achieving highly accurate and privacy-protected crowdsourced TMI without depending on the availability of massive labeled data. MTSSFL introducesconsistency updatingto insert the global model in the gradient updates of the local models that only have unlabeled data to improve their training. We also devisemean-teacher-averaging, a secure parameter aggregation mechanism that further boosts the global model’s TMI performance without requiring additional training. Our extensive case studies on a real-world data set demonstrate that MTSSFL’s classification accuracy is merely 1.1% lower than the state-of-the-art semisupervised TMI approach while being the only one to satisfy FL’s privacy-preserving constraints. In addition, MTSSFL can achieve high accuracy with less training overhead due to the proposed semisupervised learning design.

The ongoing roll-out of cloud–edge computing and Internet of Things (IoT) has been simulating the boom of new advance reservation (AR) services, such as bulk-data migration and virtual machine backup, driving the development of substrate elastic optical networks (EONs). These AR requests are initial-delay-insensitive if they are guaranteed to be completed before a predefined deadline. Therefore, the routing, modulation, and spectrum assignment (RMSA) problem is extended to the time-spectrum domain rather than the single spectrum domain. Traditional heuristic RMSA algorithms follow static procedures under handcrafted rules and assumptions, and thus cannot be optimized automatically. To solve this problem, we propose a deep reinforced deadline-driven allocation (DRDA) algorithm. To the best of our knowledge, this work is the first to leverage deep reinforcement learning (DRL) methods to solve the AR resource allocation problem. Moreover, compared with the single experiment scenario of many existing works, the DRDA algorithm is evaluated in both static and dynamic scenarios. Simulation results show that our DRDA algorithm outperforms the other leading algorithms in both static scenario and dynamic scenario.

Abstract Deep-learning based natural language processing (NLP) models are proven vulnerable to adversarial attacks. However, there is currently insufficient research that studies attacks to neural machine translations (NMTs) and examines the robustness of deep-learning based NMTs. In this paper, we aim to fill this critical research gap. When generating word-level adversarial examples in NLP attacks, there is a conventional trade-off in existing methods between the attacking performance and the amount of perturbations. Although some literature has studied such a trade-off and successfully generated adversarial examples with a reasonable amount of perturbations, it is still challenging to generate highly successful translation attacks while concealing the changes to the texts. To this end, we propose a novel Hybrid Attentive Attack method to locate language-specific and sequence-focused words, and make semantic-aware substitutions to attack NMTs. We evaluate the effectiveness of our attack strategy by attacking three high-performing translation models. The experimental results show that our method achieves the highest attacking performance compared with other existing attacking strategies.

Today, cloud storage becomes one of the critical services, because users can easily modify and share data with others in cloud. However, the integrity of shared cloud data is vulnerable to inevitable hardware faults, software failures or human errors. To ensure the integrity of the shared data, some schemes have been designed to allow public verifiers (i.e., third party auditors) to efficiently audit data integrity without retrieving the entire users’ data from cloud. Unfortunately, public auditing on the integrity of shared data may reveal data owners’ sensitive information to the third party auditor. In this paper, we propose a new privacy-aware public auditing mechanism for shared cloud data by constructing a homomorphic verifiable group signature. Unlike the existing solutions, our scheme requires at leasttgroup managers to recover a trace key cooperatively, which eliminates the abuse of single-authority power and provides non-frameability. Moreover, our scheme ensures that group users can trace data changes through designated binary tree; and can recover the latest correct data block when the current data block is damaged. In addition, the formal security analysis and experimental results indicate that our scheme is provably secure and efficient.

Recently, mobile social cloud (MSC), formed by mobile users with social ties, has been advocated to allocate tasks of big data applications instead of relying on the conventional cloud systems. However, due to the dynamic topology of networks and social features of users, how to optimally allocate tasks to mobile users based on the trust becomes a new challenge. Therefore, this paper proposes a novel incentive scheme based on the trust of mobile users in the MSC to allocate the tasks of big data. First, a social trust degree is defined according to the social tie among users, the importance of task, and the available resources of networks. With the social trust degree, the task owner can select a group of mobile users as the candidates for task allocation. Second, a reverse auction game model is developed to study the interactions among the task owner and the candidates. With the reverse auction game model, the optimal strategy of task allocation can be obtained with a low cost for the task owner where the selected candidate of mobile users can also obtain the high profit. Finally, simulation experiments are carried out to prove that the proposal can outperform other existing methods with a low delay and a high efficiency to allocate tasks in the MSC.

Network traffic classification has become an important part of network management, which is beneficial for achieving intelligent network operation and maintenance, enhancing the network quality of service (QoS), and for network security. Given the rapid development of various applications and protocols, more and more encrypted traffic has emerged in networks. Traditional traffic classification methods exhibited the unsatisfied performance since the encrypted traffic is no longer in plain text. In this work, we modeled the time-series network traffic by the recurrent neural network (RNN). Moreover, the attention mechanism was introduced for assisting network traffic classification in the form of the following two models, the attention aided long short term memory (LSTM) as well as the hierarchical attention network (HAN). Finally, relying on the ISCX VPN-NonVPN dataset, extensive experiments were conducted, showing that the proposed methods achieved 91.2 percent in accuracy while the highest accuracy of other methods was 89.8 percent relying on the same dataset.

Software-Defined Networking (SDN) and Network Functions Virtualization (NFV) technologies facilitate deploying Service Function Chains (SFCs) at clouds in efficiency and flexibility. However, it is still challenging to efficiently chain Virtualized Network Functions (VNFs) in overlay networks without knowledge of underlying network configurations. Although there are many deterministic approaches for VNF placement and chaining, they have high complexity and depend on state information of substrate networks. Fortunately, Reinforcement Learning (RL) brings opportunities to alleviate this challenge as it can learn to make suitable decisions without prior knowledge. Therefore, in this article, we propose an RL approach for efficient SFC provision in overlay networks, where the same VNFs provided by multiple vendors are with different performance. Specifically, we first formulate the problem into an Integer Linear Programming (ILP) model for benchmarking. Then, we present the online SFC path selection into a Markov Decision Process (MDP) and propose a corresponding policy-gradient-based solution. Finally, we evaluate our proposed approach with extensive simulations with randomly generated SFC requests and a real-world video streaming dataset, and implement an emulation system for feasibility verification. Related results demonstrate that performance of our approach is close to the ILP-based method and better than deep Q-learning, random, and load-least-greedy methods.

To accommodate heterogeneous and sophisticated network services, Network Function Virtualization (NFV) is invented as a hopeful networking technology. The most distinct feature of NFV is that it separates network functions from physical hardware. In the NFV architecture, various types of Virtual Network Functions (VNFs) are placed on specific software-based middleboxes by telecom providers. Traffic traverses through a sequence of Virtual Network Functions (VNFs) in pre-defined order, which is named as Service Function Chain (SFC). However, how to effectively place VNFs at different locations and steer SFC requests while minimizing energy consumption is still an open problem. Accordingly, we investigate on the joint optimization of VNF placement and traffic steering for energy efficiency in telecom networks. We first present the power consumption model in NFV-enabled telecom networks, and then formulate the studied problem as an Integer Linear Programming (ILP) model. Since the problem is proved as NP-hard, we design a polynomial algorithm that can achieve near-optimal performances based on the Markov approximation technique. In addition, our algorithm can be extended to an online version to serve dynamic arriving SFC requests. The online algorithm achieves a near-optimal long-term averaged performance. Extensive simulation results show that compared with the benchmark algorithms, in the offline and online scenario, our algorithm can reduce up to 14.08 and 13.72 percent power consumption in telecom networks, respectively.

Cloud storage services give users a great facility in data management such as data collection, storage and sharing, but also bring some potential security hazards. An utmost importance is how to ensure the integrity of data files stored in the cloud, particular for user groups without trusted managers. Existing literature focuses on integrity checking for groups with managers who have lots of permissions. To overcome the shortage of public auditing for non-manager user groups in clouds, we develop a novel framework IPANM that integrates$(t,n)$threshold technology, blinding technology, and incentive mechanism to realize an incentive privacy-preserving public auditing scheme. In IPANM, the data integrity is guaranteed by our$(t,n)$threshold signature based public auditing and the data privacy during public auditing is protected by the blinding technology. The generation of signatures can be accelerated by our blockchain-aided incentive mechanism that mobilizes the initiative of signers in the signature generation by rewarding the contributed signers. We formally prove the security of our IPANM and conduct numerical analysis and evaluation study to validate its high efficiency. The experimental results demonstrate that IPANM has lower overheads of storage, communication, and computation as compared to the state-of-the-art technique IAID-PDP and NPP.

With the growing popularity of the Internet-of-Things (IoT), a massive amount of purpose-specific, heterogeneous, inexpensive devices have been deployed. To allow these devices to perform their duties and collaborate efficiently, designing a secure and dependable communication channel is necessary. Pairing, as the fundamental procedure for establishing a trustworthy communication channel, has received extensive attention from security researchers. Previous secure pairing protocols depend on auxiliary peripherals (e.g., displays, speakers) to share the secret message, while for those products featuring with low-price, manufacturers would probably adopt insecure pairing methods to reduce the cost, so the devices may be subject to various attacks. To mitigate such a situation, we design a peripheral-free secure pairing protocol, termed SwitchPairing. Our protocol only requires users to connect the pre-pairing devices to the same power source, then randomly presses and releases the switch to generate a shared secret. It does not require additional peripherals and can defense eavesdropping and replay attacks innately. We implement a prototype via two CC2640R2F development boards and invite volunteers to participate in the experiments about bench-marking security and usability. The result of our experiments show that our protocol can fulfill the security and efficient requirement of various IoT applications.

Backdoor attacks cause model misbehaving by first implanting backdoors in deep neural networks (DNNs) during training and then activating the backdoor via samples with triggers during inference. The compromised models could pose serious security risks to artificial intelligence systems, such as misidentifying ‘stop’ traffic sign into ‘80km/h’. In this paper, we investigate the connection characteristic between the backdoor and the trigger in DNNs and observe the fact that the backdoor is implanted via establishing a link between a cluster of neurons, representing the backdoor, and the triggers. Based on this observation, we design LinkBreaker, a new generic scheme for defending against backdoor attacks. In particular, LinkBreaker deploys a neuron consistency check mechanism for identifying compromised neuron set related to the trigger. Then, the LinkBreaker regulates the model to make predictions based on benign neuron set only and thus breaks the link between the backdoor and the trigger. Compared to previous defenses, LinkBreaker offers a more general backdoor countermeasure that is not only effective against input-agnostic backdoors but also source-specific backdoors, which the later can not be defeated by majority of state-of-the-arts. Besides, LinkBreaker is robust against adversarial examples, which, to a large extent, provides a holistic defense against adversarial example attacks on DNNs, while almost all current backdoor defenses do not have such consideration and capability. Extensive experimental evaluations on real datasets demonstrate that LinkBreaker is with high efficacy of suppressing trigger inputs while incurring no noticeable accuracy deterioration on benign inputs.

Attribute-based encryption scheme is a promising mechanism to realize one-to-many fine-grained access control which strengthens the security in cloud computing. However, massive amounts of data and various data sharing requirements bring great challenges to the complex but isolated and fixed access structures in most of the existing attribute-based encryption schemes. In this paper, we propose an attribute-based hierarchical encryption scheme with extendable policy, called Extendable Hierarchical Ciphertext-Policy Attribute-Based Encryption (EH-CP-ABE), to improve the data sharing efficiency and security simultaneously. The scheme realizes the function of hierarchical encryption, in which, data with hierarchical access control relationships could be encrypted together flexibly to improve the efficiency. The scheme also achieves external and internal extension of the access structure to further encrypt newly added hierarchical data without updating the original ciphertexts or with only a minor update depending on the data sharing requirements, which simplifies the encryption process and greatly reduces the computation overhead. We formally prove the security of the scheme is IND-CCA secure in the random oracle model based on bilinear Diffie-Hellman assumption, and we also implement our scheme to demonstrate its efficiency and practicality.

A Covert Electricity-Theft Cyberattack Against Machine Learning-Based Detection Models

Internet of Things (IoT) anomaly detection is significant due to its fundamental roles of securing modern critical infrastructures, such as falsified data injection detection and transmission line faults diagnostic in smart grids. Researchers have proposed various detection methods fostered by machine learning (ML) techniques. Federated learning (FL), as a promising distributed ML paradigm, has been employed recently to improve detection performance due to its advantages of privacy-preserving and lower latency. However, existing FL-based methods still suffer from efficiency, robustness, and security challenges. To address these problems, in this article, we initially introduce a blockchain-empowered decentralized and asynchronous FL framework for anomaly detection in IoT systems, which ensures data integrity and prevents single-point failure while improving the efficiency. Further, we design an improved differentially private FL based on generative adversarial nets, aiming to optimize data utility throughout the training process. To the best of our knowledge, it is the first system to employ a decentralized FL approach with privacy-preserving for IoT anomaly detection. Simulation results on the real-world dataset demonstrate the superior performance from aspects of robustness, accuracy, and fast convergence while maintaining high level of privacy and security protection.

The low-rank matrix completion problem is fundamental in both machine learning and computer vision fields with many important applications, such as recommendation system, motion capture, face recognition, and image inpainting. In order to avoid solving the rank minimization problem which is NP-hard, several surrogate functions of the rank have been proposed in the literature. However, the matrix restored from the optimization problem based on the existing surrogate functions seriously deviates from the original one. In this paper, we first design a new non-convex Schatten capped$p$norm which generalizes several existing non-convex matrix norms and balances between the rank and the nuclear norm of the matrix. Then, a matrix completion method based on the Schatten capped$p$norm is proposed by exploiting the framework of the alternating direction method of multipliers. Meanwhile, the Schatten capped$p$norm regularized least squares subproblem is analyzed in detail and is solved explicitly. Finally, we evaluate the performance of the proposed matrix completion method based on extensive experiments in the field of image inpainting. All the experimental results demonstrate that the proposed method can indeed improve the accuracy of matrix completion compared with the existing methods.

The UHF radio-frequency identification (RFID) has gained growing attention for tagged object localization in smart storage systems. Due to Non-Line-Of-Sight (NLOS) condition, it is challenging to accurately locate the position of tags inside closed spaces. In this paper, we propose a precise and cost-effective solution for tagged object localization in closed spaces, using only received signal strength (RSS) information. We establish a RSS profile for each tag and discover some important features of RSS profiles including uniqueness, time-variation, column-dependence and waveform-similarity. Based on these features, we propose a reference-free RSS-profile (RFRP) localization scheme. The advantage of our propose scheme is to accurately localize multiple tags in closed spaces by overcoming the challenges including the lack of pre-deployed reference tags, NLOS propagation, multi-path propagation and coupling effect. The RFRP scheme first roughly estimates tags’ coordinates based on Peak Asymmetry Factor, then acquires reference-tag substitutes through the similarity of RSS sequences. Subsequently, our scheme refines the relative positions of all tags by these substitutes. Finally all tags’ absolute positions are estimated through a RSS-ranging model. Extensive experiment results demonstrate that our approach can achieve high ordering accuracy and localization accuracy for the tags inside closed spaces.

Subscription-based services, where broadcasting companies make profits by providing subscription programs to users, have become a popular application of wireless broadcast systems. Driving by business interests and data security, companies usually encrypt content of programs with session keys, then only authorized users can access the contents of their subscribed programs, respectively. It is challenging to flexibly distribute and update the session key, we are therefore motivated to propose an identity-based many-to-many subscription scheme (IMS) with efficient key management. The IMS scheme possesses three novel advantages: First, it provides a whole set of subscription flow, from the selection of programs of interest to the generation and updating of keys. Second, it supports program updating and all subscription behaviors. Third, it employs a signature and authentication mechanism to stipulate that only authorized users can access the programs’ ciphertext. The theoretical analysis demonstrates that the proposed scheme is secure against chosen-ciphertext attacks (CCA) under the$ k$-bilinear Diffie-Hellman exponent (BDHE) problem, and has a decent performance in storage and computational overhead. The experimental analysis shows the total time cost of the proposed scheme is reduced more than 75.02 percent compared against the popular peer schemes under the condition of 2500 users.

Virtual network embedding (VNE) can effectively deploy virtual networks (VNs) onto shared substrate network (SN) resources. However, with the consistent changing scalability and diversity demands of VNs, traditional VNE methods prove to be a challenging task for current cloud service platforms. Thus, we model a repeatable multi-dimensional virtual network embedding (RMD-VNE) problem for implementing multi-dimensional virtual networks (MD-VNs) that involves real servers, virtual machines, containers, and network simulators. The MD-VN is preprocessed and embedded via a heuristic method denoted asReMiDvne. Following its transformation for the containers and simulation networks, the MD-VN topology undergoes a process of coarsening, partitioning, and uncoarsening.ReMiDvnethen applies a topology-aware repeatable embedding solution to complete the embedding stage. Experimental results demonstrate thatReMiDvneoutperforms seven baseline approaches through small-, 1,000- and 10,000-scale VNE simulation experiments. Remarkably,ReMiDvneimproves the average rates of acceptance ratio, revenue, and revenue-cost ratio by up to 40.45, 40.45, and 299.03 percent, respectively, and reduces the average rate of cost by up to 64.16 percent. Furthermore, real-world VNE experiments are conducted based on the OpenStack platform. The results reveal the ability ofReMiDvneto efficiently reduce communication costs by up to 45.93 and 63.43 percent for download and upload, respectively.

Space-air-ocean integrated networks (SAOINs), composed of low earth orbit (LEO) satellites, unmanned aerial vehicles (UAVs), and unmanned surface vehicles (USVs), have been advocated to provide seamless, high-rate, and reliable wireless transmission services for USVs. However, due to the restrictions of limited resources (e.g., spectrum bandwidth, transmission power, etc.), diverse demands of USVs, and selfishness of both UAVs and LEOs, there comes a significant challenge to provision high-quality wireless data rate for USVs to achieve their satisfied quality of experience (QoE). To this end, in this paper, we propose a hierarchical on-demand wireless data rate provisioning scheme to provide ubiquitous transmission services for USVs. Specifically, we first devise a hierarchical wireless data rate provisioning framework. The LEO satellite with an extensive wireless coverage is utilized to provide LEO satellite-to-UAV (L2U) data rate for UAVs with a certain L2U data rate price. Each UAV is employed to provide UAV-to-USV (U2U) data rate for covered multiple USVs with a certain U2U data rate price. We then propose a modified three-stage Stackelberg game to model the wireless data rate assignments among LEO satellites, UAVs, and USVs, where the time-varying data rate demands of USVs are considered to formulate the utility maximization problem. Afterwards, the backward induction approach is leveraged to attain the Stackelberg equilibrium as the solution of the formulated problem, where the closed-form expressions on the optimal strategies of both USVs and UAVs under different data rate budgets are obtained by the nonlinear programming method. Besides, an accelerated conjugate gradient descent (ACGD) based iteration algorithm is also designed to obtain the optimal strategies of the LEO satellites on the L2U data rate prices. At last, extensive simulations are carried out to demonstrate that the proposed scheme can significantly increase the utilities of USVs, as compared to other benchmark schemes.

In today's “Internet of Everything (IoE)” era, the collaboration from massive participants significantly boosts the performance and efficiency of model training. This trend also un-avoidably stirs up considerable concerns about multi-dimensional security problems. Under the circumstances, federated learning (FL) is enthusiastically adopted, as it protects privacy to a certain extent by only processing personal data locally. Nevertheless, FL's characteristics of concealment also pave the way for sev-eral emerging attacks during the training process, i.e., model inversion, poisoning, and backdoor. Currently, although partially mitigating attack effects, existing countermeasures against those threats are studied separately and orthogonal. This separation makes those defense methods mutually exclusive and restrictive in real-world application scenarios, far from satisfying. In this paper, we extensively model different attack paradigms into three types based on CIA-triad, the well-known information security primitive, and propose a novel dual-shuffle method to thwart aforementioned threats jointly. Concretely speaking, our primary model shuffling mechanism provides the confidentiality guarantee based on the information-theoretic notion of identifiability; then, an augmented client shuffling mechanism purges the user group of adversaries proactively without any compromise of anonymous constraints. By conducting a series of experiments on bench-mark datasets, we demonstrate that our method could achieve significant security and convergence performance against three state-of-the-art attacks.

The software-defined Low Earth Orbit (LEO) satellite network can effectively improve the flexibility of the inter-satellite networking. However, the impact of Inter-Satellite Link (ISL) attributes on the network topology has not been well investigated in existed works, which leads to the loss of topology information and the unreliability of routing path. In this paper, we propose a dynamic routing algorithm based on ISL attributes to improve the adaptability and reliability of routing in LEO satellite network. Firstly, we construct the utility function of ISL attributes to quantify the link utility. Then, the maximum deviation algorithm is introduced to adaptively calculate the weight of attribute parameters in link utility function, which can quantify the impact of each attribute on link quality. Finally, the mathematical model of path utility is established, the routing path with highest path utility is selected as the optimal routing path. The simulation results show that the proposed scheme can improve the network performance by about 35%, 10% and 20% in terms of packet drop rate, end-to-end delay and throughput.

Loading the functions of base station on satellite can improve the flexibility of user access and expand the effective coverage of Satellite-Terrestrial Integrated Network (STIN). However, the functions of management and control that supports the satellite networking have not been well investigated. A large amount of signaling needs to be forwarded to the ground control center for processing, which increases the delay of network control and management. In this paper, we propose a novel mobile core network architecture that loads the mobility management function of core network on the satellite, which can improve the flexibility of management and control of STIN. Firstly, considering the on-board computing capacity and the dynamic characteristic of satellites, we design the lightweight Satellite Mobile Core Network (SMCN) to improve the flexibility of the Satellite next-generation NodeB (Sat-gNB) networking. Then, the interactive protocol for user access control, mobility control of Sat-gNBs and the coordinated control between ter-restrial network and satellite network are designed to support the mobility management. Finally, in order to optimize the delay of Sat-gNBs networking, we construct the mathematical model for the multi-SMCN deployment. The simulation results show that compared to the architecture of Non-Terrestrial Networks (NTN), the proposed mobile core network architecture can improve the flexibility of mobility management and optimize the network delay.

In federated learning, more and more studies have discovered that attackers can recover the original data from the shared gradients of participants. However, existing defense models struggle to balance the privacy of participants and the effectiveness of federated learning in the face of cutting-edge attack models. Therefore, we propose a powerful defense model to protect the original data while ensuring the effect of classification. First, we get two featured datasets from original data based on Sparse Dictionary Learning (DL) or QR decomposition. In these two featured datasets, we select one dataset to replace the original data for federated training named co-trained data, and the other one is kept on the local client named left data. At this point, the adversary in federated learning can only obtain co-trained data, which cannot recover the original data due to the lack of left data. Following the completion of the federated learning, the participant requests a parameter from the server. Using this parameter, we can combine the aggregated global model over co-trained data with the offline-trained local model of an arbitrary participant to develop the final classification results. Some theories and a lot of experiments demonstrate the classification effectiveness of our model. It also can be a general solution to the original data leakage problems caused by gradient leakage.

The unprecedented growth of mobile data traffic brings unique challenges for network bandwidth and server resources to meet the diverse QoE (Quality of Experience). Caching becomes a promising way to alleviate these issues by storing a subset of data at the network edge, for which caching policy becomes critical. To this end, various caching schemes have been put forward, however, these schemes are either not intelligent lacking the ability of self-learning and self-decision-making, or inefficient with low data hit rate. Based on these observations, in this paper, we propose a novel Intelligent Caching framework at the Edge, named ICE, via deep reinforcement learning to capture certain valued information of the requested data. Notably, in our approach, the popularity of the data to be cached will be explored and considered. A Markov decision model is further developed to determine whether the data should be cached. The evaluation shows that ICE greatly improves the hit rate in comparison with the state-of-the-art approaches, and reduces the energy consumption for data transmission. Furthermore, based on ICE, the users' QoE is greatly improved. In conclusion, both theoretical analysis and experimental results prove the effectiveness and high performance of ICE compared with conventional strategies.

Machine learning as a service (MLaaS) brings many benefits to people's daily life. However, the service mode of MLaaS will increase the risk of users' privacy leakage. Existing works focusing on privacy-preserving based on encryption, differential privacy, and distributed framework require high computing resources or cannot be applied in MLaaS. In this paper, we propose feature dilution (FD), a noise-based desensitization algorithm to remove sensitive information in raw data. In particular, FD continuously adds raw data features to the random noise until it meets the minimum amount for an effective query, and we call this noise weak-feature noise (WFN). By fine-tuning the MLaaS architecture, we have realized that users can utilize WFN to get normal services without exposing their local private data. Meanwhile, noise addition technology is introduced by us to reduce the risk of privacy leakage caused by “weak features”. Extensive experiments have demonstrated that users can use FD to obtain effective services without exposing their private data. Finally, we conducted practical tests on weak-feature noises and found that these noises are difficult to use by malicious service providers.

Nowadays, the quality of cloud services offered by different service providers varies greatly. Reputation mechanism, as a better service evaluation method, can help standardize and regulate the cloud service market. However, existing reputation systems either rely on a trusted third party with security and privacy issues, or lack reliable evaluation. To address the above issues, we propose a blockchain based privacy-preserving dynamic reputation mechanism for cloud service and a reputation management smart contract (RM) is designed to implement trusted reputation computation. The reputation integrates conformance trust in the subjective view and recommendation trust in the objective view together to provide a comprehensive evaluation. Besides, a miner selection algorithm is designed to prevent miners from launching a collusion attack. Moreover, the Paillier homomorphic encryption algorithm (PHE) is introduced to encrypt the sensitive data of customers, ensuring the security of data stored on the blockchain. Experiment results reveal that the proposed model is feasible and the performance of encryption is acceptable.

The distributed denial of service (DDoS) attack is detrimental to businesses and individuals as people are heavily relying on the Internet. Due to remarkable profits, crackers favor DDoS as cybersecurity weapons to attack a victim. Even worse, edge servers are more vulnerable. Current solutions lack adequate consideration to the expense of attackers and inter-defender collaborations. Hence, we revisit the DDoS attack and defense, clarifying the advantages and disadvantages of both parties. We further propose a joint defense framework to defeat attackers by incurring a significant increment of required bots and enlarging attack expenses. The quantitative evaluation and experimental assessment showcase that such expense can surge up to thousands of times. The skyrocket of expenses leads to heavy loss to the cracker, which prevents further attacks.

Machine learning has been increasingly adopted to detect Android malwares. Most existing studies depend on features in code space such as information flows and API calls. Malware variants would engage these models in a never-ending war. Inspired by the observation that some variants share similar or even identical user interfaces (UIs), this paper explores employing visible UI screenshot as the indicator to build a novel Android malware detection method. To achieve this vision, we built the first Android Application Screenshot Dataset (AnASD) consisting of more than twenty thousand UI screenshots produced by both benign applications and malwares. A thorough analysis was conducted to characterize the dataset, especially the UI difference between benign applications and malwares. Then a set of state of the art deep learning classifiers on AnASD were trained and evaluated. The results of both sim-ilarity measurement and classification performance proved the feasibility to detect Android malwares based on user interfaces. To facilitate the research community, the dataset is free available at https://doi.org/10.6084/m9.figshare.14445768.

With the development of network, network security has become a topic of increasing concern. Recent years, machine learning technology has become an effective means of network intrusion detection. However, machine learning technology requires a large amount of data for training, and training data often contains privacy information, which brings a great risk of privacy leakage. At present, there are few researches on data privacy protection in the field of intrusion detection. Regarding the issue of privacy and security, we combine differential privacy and machine learning algorithms, including One-class Support Vector Machine (OCSVM) and Local Outlier Factor(LOF), to propose an hybrid intrusion detection system (IDS) with privacy protection. We add Laplacian noise to the original network intrusion detection data set to get differential privacy data sets with different privacy budgets, and proposed a hybrid IDS model based on machine learning to verify their utility. Experiments show that while protecting data privacy, the hybrid IDS can achieve detection accuracy comparable to traditional machine learning algorithms.

To enhance dependable resource allocation against increasing distributed denial-of-service (DDoS) attacks, in this article, we investigate interactions between a sensor device-edgeVM pair and a DDoS attacker using a game-theoretic framework, under the constraints of the task time, resource budget, and incomplete knowledge of the processing time of machine learning tasks. In this game, the sensor device expects an edgeVM to cooperate and choose its resource allocation strategy with the objective of satisfying the minimum resource required of machine learning tasks at the corresponding sensor device. Similarly, the attacker's objective is to strategically allocate resources so that the resource constraint of the machine learning tasks is not satisfied. Owing to a lack of complete information of the processing time of the machine learning tasks, this strategic resource allocation problem between the two players is modeled as a Bayesian Q-learning game, in which the optimal strategies of the sensor device-edgeVM pair and the attacker are analyzed. Furthermore, probability distributions are employed by the corresponding players to model the incomplete nature of the game and a greedy Q-learning algorithm is proposed to dependable resource allocation against DDoS attacks. Numerical simulation results demonstrate that the proposed mechanism is superior to other dependable resource allocation mechanisms under incomplete information for DDoS attacks in the sensor edge cloud.

With the advent of the Internet of Things (IoT) and the increasing popularity of the intelligent transportation system, a large number of sensing devices are installed on the road for monitoring traffic dynamics in real time. These sensors can collect streaming traffic data distributed across different traffic sites, which constitute the main source of big traffic data. Analyzing and mining such big traffic data in massive IoT can help traffic administrations to make scientific and reasonable traffic scheduling decisions, so as to avoid prospective traffic congestions in the future. However, the above traffic decision making often requires frequent and massive data transmissions between distributed sensors and centralized cloud computing centers, which calls for lightweight data integrations and accurate data analyses based on large-scale traffic data. In view of this challenge, a big data-driven and nonparametric model aided by 6G is proposed in this article to extract similar traffic patterns over time for accurate and efficient short-term traffic flow prediction in massive IoT, which is mainly based on time-aware locality-sensitive hashing (LSH). We design a wide range of experiments based on a real-world big traffic data set to validate the feasibility of our proposal. Experimental reports demonstrate that the prediction accuracy and efficiency of our proposal are increased by 32.6% and 97.3%, respectively, compared with the other two competitive approaches.

User profile perturbation protects privacy in the release of user profiles to receive recommendation services, in which the privacy budget as a privacy parameter can be controlled to effect a tradeoff between the recommendation quality and privacy protection against inference attacks. In this article, we propose a deep reinforcement learning (RL)-based user profile perturbation scheme for recommendation systems. This scheme applies differential privacy to protect user privacy and uses deep RL to choose the privacy budget against inference attackers. Based on an evaluated neural network (NN) and a target NN, this scheme enables a user device to optimize the privacy budget over time based on the sensitivity level of the clicked item, the similarities among the recommended items, and the estimated privacy loss. We provide an upper bound on the privacy protection performance of this scheme in the recommendation game and evaluate its computational complexity. Simulation results for a movie recommendation system show that this scheme increases the user privacy protection level for a given recommendation quality compared with benchmark schemes.

The infectious disease presents great hazards to public health, due to their high infectivities and potential lethalities. One of the effective methods to hinder the spread of infectious disease is vaccination. However, due to the limitation of resource and the medical budget, vaccinating all people is not feasible in practice. Besides, the vaccinating effects are difficult to be timely observed through traditional ways, such as outpatient services. To tackle the above problems, we propose an e-healthcare mobile social Internet of Things (MSIoTs)-based targeted vaccination scheme to fast contain the spread of the infectious disease. Specifically, we first develop an e-healthcare MSIoT architecture by integrating the e-healthcare system and MSIoTs, whereby the spread status of the infectious disease is timely collected. Furthermore, a graph coloring and spreading centrality-based optional candidate searching algorithm is devised to hunt for the candidates that are powerfully capable of preventing infectious disease. Especially, in order to reduce the vaccination cost, we design an optimal vaccinated target selection algorithm to choose a minimum number of targets whose locations are differentially distributed. Extensive simulations demonstrate that the proposed scheme can effectively prevent infectious disease as compared to conventional schemes.

Moving target defense (MTD) is an emerging proactive defense technology, which can reduce the risk of vulnerabilities exploited by attacker. As a crucial component of MTD, route mutation (RM) faces a few fundamental problems defending against sophisticated Distributed-Denial of Service (DDoS) attacks: 1) it is unable to make optimal mutation selection due to insufficient learning in attack behaviors and 2) because network situation is time varying, RM also lacks self-adaptation in mutation parameters. In this article, we propose a context-aware Q-learning algorithm for RM (CQ-RM) that can learn attack strategies to optimize the selection of mutated routes. We first integrate four representative attack strategies into a unified mathematical model and formalize multiple network constraints. Then, taking above network constraints into considerations, we model RM process as a Markov decision process (MDP). To look for the optimal policy of MDP, we develop a context estimation mechanism and further propose the CQ-RM scheme, which can adjust learning rate and mutation period adaptively. Correspondingly, the optimal convergence of CQ-RM is proved theoretically. Finally, extensive experimental results highlight the effectiveness of our method compared to representative solutions.

Edge computing is a key-enabling technology that meets continuously increasing requirements for the intelligent Internet-of-Things (IoT) applications. To cope with the increasing privacy leakages of machine learning while benefiting from unbalanced data distributions, federated learning has been wildly adopted as a novel intelligent edge computing framework with a localized training mechanism. However, recent studies found that the federated learning framework exhibits inherent vulnerabilities on active attacks, and poisoning attack is one of the most powerful and secluded attacks where the functionalities of the global model could be damaged through attacker's well-crafted local updates. In this article, we give a comprehensive exploration of the poisoning attack mechanisms in the context of federated learning. We first present a poison data generation method, named Data_Gen, based on the generative adversarial networks (GANs). This method mainly relies upon the iteratively updated global model parameters to regenerate samples of interested victims. Second, we further propose a novel generative poisoning attack model, named PoisonGAN, against the federated learning framework. This model utilizes the designed Data_Gen method to efficiently reduce the attack assumptions and make attacks feasible in practice. We finally evaluate our data generation and attack models by implementing two types of typical poisoning attack strategies, label flipping and backdoor, on a federated learning prototype. The experimental results demonstrate that these two attack models are effective in federated learning.

Research has shown the potential of device-free WiFi sensing for human identification. Each and every human has a unique gait and prior works suggest WiFi devices are able to capture the unique signature of a person's gait. In this article, we show for the first time that the monitored gait could be inconsistent and have mirror-like perturbations when individuals walk through WiFi devices in different directions, provided that the WiFi antenna array is horizontal to the walking path. Such inconsistent mirrored patterns are to negatively affect the uniqueness of gait and accuracy of human identification. Therefore, we propose a system called Gate-ID for accurately identifying individuals' identities irrespective of different walking directions. Gate-ID employs theoretical communication model and real measurements to demonstrate that antenna array orientations and walking directions contribute to the mirror-like patterns in WiFi signals. A novel heuristic algorithm is proposed to infer individual's walking directions. A set of methods are employed to extract and augment the representative spatial-temporal features of gait and enable the system performing irrespective of walking directions. We further propose a novel attention-based deep learning model that fuses various weighted features and ignores ineffective noises to uniquely identify individuals. We implement Gate-ID on commercial off-the-shelf devices. Extensive experiments demonstrate that our system can uniquely identify people with average accuracy of 90.7%-75.7% from a group of 6-20 people, respectively, and improve the accuracy by 12.5%-43.5% compared with baselines.

Cloud-fog computing emerges to satisfy the low latency and high computation requirements of Internet of Things (IoT) services. Elastic optical networks (EONs) are excellent substrate communication networks between fog datacenters and cloud datacenters. However, the uneven traffic of massive cloud-fog services incurs many spectrum fragments, leading to high extra energy consumption. To solve this problem, we propose an energy-efficient deep reinforced traffic grooming (EDTG) algorithm based on deep reinforcement learning. Unlike existing manually network features extracting methods, we convert the traditional network modal and the service routing path into colored network images to represent their states and extract the features automatically by MobilenetV3 according to these images. With the extracted features, we implement an advantage actor-critic (A2C) algorithm, whose actor module and critic module share an artificial neural network (ANN) to get optimal grooming actions. Additionally, after repeated attempts and experiments, we set up an objective reward and punishment mechanism to evaluate the grooming actions. We conduct extensive simulations for performance evaluation, and the results have shown that EDTG can significantly reduce energy consumption compared with two well-performed traffic grooming algorithms.

Digital humanities is an important subject because it enables developments in history, literature, and films. In this article, we perform an empirical study of a Chinese historical text, Records of the Three Kingdoms (Records), and a historical novel of the same story, Romance of the Three Kingdoms (Romance). We employ deep-learning-based natural language processing (NLP) techniques to extract characters and their relationships. The adopted NLP approach can extract 93% and 91% characters that appeared in the two books, respectively. Then, we characterize the social networks and sentiments of the main characters in the historical text and the historical novel. We find that the social network in Romance is more complex and dynamic than that of Records, and the influence of the main characters differs. These findings shed light on the different styles of storytelling in the two literary genres and how the historical novel complicates the social networks of characters to enrich the literariness of the story.

Next point-of-interest (POI) recommendation has been an important task for location-based intelligent services. However, the application of such promising technique is still limited due to the following three challenges: 1) the difficulty of capturing complicated spatiotemporal patterns of user movements; 2) the hardness of modeling fine-grained long-term preferences of users; and 3) the effective learning of interaction between long- and short-term preferences. Motivated by this, we propose a memory augmented hierarchical attention network (MAHAN), which considers both short-term check-in sequences and long-term memories. To capture the complicated interest tendencies of users within a short-term period, we design a spatiotemporal self-attention network (ST-SAN). For long-term preferences modeling, we employ a memory network to maintain fine-grained preferences of users and dynamically operate them based on users' constantly updated check-ins. Moreover, we first employ a coattention network/mechanism to integrate the proposed ST-SAN and memory network, which can fully learn the dynamic interaction between long- and short-term preferences. Our extensive experiments on two publicly available data sets demonstrate the effectiveness of MAHAN.

The Internet of Things (IoT) is becoming truly ubiquitous in every domain of human lives, and a large number of objects can be connected and enabled to communicate with cloud servers at any time. However, complex connections and vulnerabilities of IoT devices introduce inevitable security threats, in which distributed denial-of-service (DDoS) attacks usually incur catastrophic results. Unfortunately, the existing DDoS mitigation methods cannot provide effective protection. Moreover, the amplifying complexity and increasing delay incurred by defense greatly affect the stability of IoT networks. To tackle these problems, we present a novel framework that can proactively adapt the attack surface of IoT networks, dynamically optimize defense strategies, and rapidly deploy the corresponding defense mechanisms. In particular, we establish hybrid proactive defense mechanisms combining Moving Target Defense (MTD) techniques with cyber deception to spread camouflage information to confuse attackers. Based on these mechanisms, we introduce a defender-led signaling game model to formalize defense scenarios and depict the interactions between the defender and the attacker. Besides, we present an optimal algorithm to solve decision problems and optimize defense implementation in a cost-effective manner. Our extensive experiments demonstrate that the proposed approach can effectively mitigate DDoS attacks and maintain a high level of performance in IoT networks with acceptable overhead.

Federated learning has been applied to various tasks in intelligent transportation systems to protect data privacy through decentralized training schemes. The majority of the state-of-the-art models in intelligent transportation systems (ITS) are graph neural networks (GNN)-based for spatial information learning. When applying federated learning to the ITS tasks with GNN-based models, the existing frameworks can only protect the data privacy; however, ignore the one of topological information of transportation networks. In this article, we propose a novel federated learning framework to tackle this problem. Specifically, we introduce a differential privacy-based adjacency matrix preserving approach for protecting the topological information. We also propose an adjacency matrix aggregation approach to allow local GNN-based models to access the global network for a better training effect. Furthermore, we propose a GNN-based model named attention-based spatial-temporal graph neural networks (ASTGNN) for traffic speed forecasting. We integrate the proposed federated learning framework and ASTGNN as FASTGNN for traffic speed forecasting. Extensive case studies on a real-world dataset demonstrate that FASTGNN can develop accurate forecasting under the privacy preservation constraint.

Software defined satellite-terrestrial networking has been identified as a promising approach to support the diversity of network services. As the fundamental issue to improve the flexibility of network management, the controller placement problem has been attracted increasing attentions for the integration of satellite and terrestrial networking. However, the impact of the dynamic coverage demands on the controller placement have not been well investigated in existed works, which makes them fail to adjust the coverage dynamically according to the actual demands, and leads to an obvious increase of networking response latency to the terminals. Aiming to address this issue, we propose a novel on-demand dynamic controller placement scheme, which can optimize the placement of controllers to improve networking response latency while meeting the dynamic coverage demands. Firstly, to optimize the number of controllers and meet the dynamic coverage demands, we define the coverage redundancy and propose the redundancy-based satellite subnet division method to establish the reliable satellite subnets. Secondly, we quantify the networking response latency of the distributed satellite subnet, and build an optimization mathematical model to optimize the number and location of controllers. Then, we formulate the controller placement problem into the capacitated facility location problem and build the mathematical model for it. Moreover, the on-demand dynamic approximation algorithm is proposed to obtain the approximation solution. Finally, the simulation results demonstrate that the proposed algorithm can effectively optimize the network latency compared with related algorithms.

With the ever-increasing prosperity of web Application Programming Interface (API) sharing platforms, it is becoming an economic and efficient way for software developers to design their interested mashups through web API re-use. Generally, a software developer can browse, evaluate, and select his or her preferred web APIs from the API's sharing platforms to create various mashups with rich functionality. The big volume of candidate APIs places a heavy burden on software developers’ API selection decisions. This, in turn, calls for the support of intelligent API recommender systems. However, existing API recommender systems often face two challenges. First, they focus more on the functional accuracy of APIs while neglecting the APIs’ actual compatibility. This then creates incompatible mashups. Second, they often require software developers to input a set of keywords that can accurately describe the expected functions of the mashup to be developed. This second challenge tests partial developers who have little background knowledge in the fields. To tackle the above-mentioned challenges, in this article we propose a compatibility-aware and text description-driven web API recommendation approach (named WAR text ). WAR text guarantees the compatibility among the recommended APIs by utilizing the APIs’ composition records produced by historical mashup creations. Besides, WAR text entitles a software developer to type a simple text document that describes the expected mashup functions as input. Then through textual description mining, WAR text can precisely capture the developers’ functional requirements and then return a set of APIs with the highest compatibility. Finally, through a real-world mashup dataset ProgrammableWeb, we validate the feasibility of our novel approach.

The demand for data center bandwidth has exploded due to the continuous development of cloud computing, causing the use of network resources close to saturation. Optical network has become an encouraging technology for many burgeoning networks and parallel/distributed computing applications because of its huge bandwidth. This article focuses on efficient embedding of data centers into optical networks, which aims to reduce complexity of the network topology by using the parallel transmission characteristics of optical fiber. We first present a novel virtual network embedding (VNE) mathematical model used for optical data center networks. Then we derive a priority of location VNE algorithm according to node proximity sensing and path comprehensive evaluation. Furthermore, we propose routing and wavelength assignment for DCNs into optical networks, and identify the lower bound of the required number of wavelengths. Extensive evaluations show that the proposed embedding algorithm can reduce the average waiting time of virtual network requests by 20 percent, increase the request acceptance rate and revenue-overhead ratio by 13 percent, as compared to the latest VNE algorithm.

Community detection for Location-based Social Networks (LBSNs) has been received great attention mainly in the field of large-scale Wireless Communication Networks. In this paper, we present a Differential Privacy Latent Trajectory cOmmunity Discovering (DP-LTOD) scheme, which obfuscates original trajectory sequences into differential privacy-guaranteed trajectory sequences for trajectory privacy-preserving, and discovers latent trajectory communities through clustering the uploaded trajectory sequences. Different with traditional trajectory privacy-preserving methods, we first partition original trajectory sequence into different segments. Then, the suitable locations and segments are selected to constitute obfuscated trajectory sequence. Specifically, we formulate the trajectory obfuscation problem to select an optimal trajectory sequence which has the smallest difference with original trajectory sequence. In order to prevent privacy leakage, we add Laplace noise and exponential noise to the outputs during the stages of location obfuscation matrix generation and trajectory sequence function generation, respectively. Through formal privacy analysis, we prove that DP-LTOD scheme can guarantee ϵ-differential private. Moreover, we develop a trajectory clustering algorithm to classify the trajectories into different kinds of clusters according to semantic distance and geographical distance. Extensive experiments on two real-world datasets illustrate that our DP-LTOD scheme can not only discover latent trajectory communities, but also protect user privacy from leaking.

In the area of community question answering (CQA), answer selection and answer ranking are two tasks which are applied to help users quickly access valuable answers. Existing solutions mainly exploit the syntactic or semantic correlation between a question and its related answers (Q&A), where the multifacet domain effects in CQA are still underexplored. In this paper, we propose a unified model, enhanced attentive recurrent neural network (EARNN), for both answer selection and answer ranking tasks by taking full advantages of both Q&A semantics and multifacet domain effects (i.e., topic effects and timeliness). Specifically, we develop a serialized long short-term memory to learn the unified representations of Q&A, where two attention mechanisms at either sentence level or word level are designed for capturing the deep effects of topics. Meanwhile, the emphasis of Q&A can be automatically distinguished. Furthermore, we design a time-sensitive ranking function to model the timeliness in CQA. To effectively trainEARNN, a question-dependent pairwise learning strategy is also developed. Finally, we conduct extensive experiments on a real-world dataset from Quora. Experimental results validate the effectiveness and interpretability of our proposedEARNNmodel.

Traffic speed prediction is among the key problems in intelligent transportation system (ITS). Traffic patterns with complex spatial dependency make accurate prediction on traffic networks a challenging task. Recently, a deep learning approach named Spatio-Temporal Graph Convolutional Networks (STGCN) has achieved state-of-the-art results in traffic speed prediction by jointly exploiting the spatial and temporal features of traffic data. Nonetheless, applying STGCN to large-scale urban traffic network may develop degenerated results, which is due to redundant spatial information engaging in graph convolution. In this work, we propose a motif-based graph-clustering approach to apply STGCN to large-scale traffic networks. By using graph clustering, we partition a large urban traffic network into smaller clusters to prompt the learning effect of graph convolution. The proposed approach is evaluated on two real-world datasets and is compared with its variants and baseline methods. The results show that graph-clustering approaches generally outperform the other methods, and the proposed approach obtains the best performance.

To ensure data security and privacy during big data publishing, it is challenging to design a security and privacy protection scheme for the big data environment with a large scale of users. At the same time, due to the users' dynamically joining and exiting, it is also very important to design a user's dynamic update mechanism. To address such challenges, we design a novel scalable and updatable attribute-based privacy protection scheme (SUAPP) for big data publishing. The proposed scheme can realize users' hierarchical management, which can reduce the overhead on key generation and management caused by the large scale of data users in the big data center (BDC). We set a user group for each attribute, then adapt the Chinese remaining theorem to dynamically assist the big data center to generate and update group keys for the attribute users group. Analyses and experiments show that while ensuring the privacy protection of big data publishing, our scheme also has low communication and computation overhead and higher efficiency compared with two state peer schemes.

The discovery of adversarial examples poses a serious risk to the deep neural networks (DNN). By adding a subtle perturbation that is imperceptible to the human eye, a well-behaved DNN model can be easily fooled and completely change the prediction categories of the input samples. However, research on adversarial attacks in the field of modulation recognition mainly focuses on increasing the prediction error of the classifier, while ignores the importance of decreasing the perceptual invisibility of attack. Aiming at the task of DNNbased modulation recognition, this study designs the Fitting Difference as a metric to measure the perturbed waveforms and proposes a new method: the Nesterov Adam Iterative Method to generate adversarial examples. We show that the proposed algorithm not only exerts excellent white-box attacks but also can initiate attacks on a black-box model. Moreover, our method decreases the waveform perceptual invisibility of attacks to a certain degree, thereby reducing the risk of an attack being detected.

Federated learning is a state-of-the-art technology used in the fog computing, which allows distributed learning to train cross-device data while achieving efficient performance. Many current works have optimized the federated learning algorithm in homogeneous networks. However, in the actual application scenario of distributed learning, data is independently generated by each device, and this non-homologous data has different distribution characteristics. Therefore, the data used by each device for local learning is unbalanced and non-IID, and the heterogeneity of data affects the performance of federated learning and slows down the convergence. In this paper, we present a dynamic sample selection optimization algorithm, FedSS, to tackle heterogeneous data in federated learning. FedSS dynamically selects the training sample size during the gradient iteration based on the locally available data size, to settle the expensive evaluations of the local objective function with a massive amount of dataset. We theoretically analyze the convergence and present the complexity estimates of our framework when learning large data from unbalanced distribution. Our experimental results show that the use of dynamic sampling methods can effectively improve the convergence speed with heterogeneous data, and keep computational costs low while achieving the desired accuracy.

Federated learning has lately received great attention for its privacy protection feature. However, recent researches found that federated learning models are susceptible to various inference attacks. In this paper, we point out a membership inference attack method that can cause a serious privacy leakage in federated learning. An adversary who is a participant in federated learning can train a classification attack model to launch the membership inference attack, which determines if a data record is in the model's training dataset. The existing membership inference method is dissatisfied due to a lack of attack data since the training data of each participant are independent. To overcome the lack of attack data, an adversary can enrich attack data using the generative adversarial network (GAN), which is a practical method to increase data diversity. We substantiate that this GAN enhanced membership inference attack method has a 98% attack accuracy. We perform experiments to show that data diversity and the overfitting make federated learning models susceptible.

Non technical losses (NTL) detection plays a crucial role in protecting the security of smart grids. Employing massive energy consumption data and advanced artificial intelligence (AI) techniques for NTL detection are helpful. However, there are concerns regarding the effectiveness of existing AI-based detectors against covert attack methods. In particular, the tampered metering data with normal consumption patterns may result in low detection rate. Motivated by this, we propose a hybrid data-driven detection framework. In particular, we introduce a wide & deep convolutional neural networks (CNN) model to capture the global and periodic features of consumption data. We also leverage the maximal information coefficient algorithm to analysis and detect those covert abnormal measurements. Our extensive experiments under different attack scenarios demonstrate the effectiveness of the proposed method.

Due to robustness against multi-path effect, channel state information (CSI) of Orthogonal Frequency Division Multiplexing (OFDM) systems is supposed to provide accurate distance measurement for indoor localization. However, we find that the original CSI ranging model is biased, so the model cannot be used to directly derive Cramer-Rao lower bound (CRLB) of positioning error for CSI-ranging based localization scheme. In this paper we first analyze the estimation bias of the original CSI ranging model according to indoor wireless channel model. Then we propose a negative power summation ranging model which can be used as an unbiased ranging model for both Line-Of-Sight (LOS) and Non-LOS scenarios. Subsequently, based on the proposed model, we derive both the CRLB of ranging error and the CRLB of positioning error for CSI-ranging localization scheme. Through simulation we validate the bias of the original ranging model and the approximately zero bias of our proposed ranging model. Through comprehensive experiments in different indoor scenarios, localization errors by different ranging models are compared to the CRLB, meanwhile our proposed ranging model is demonstrated to have better ranging and localization accuracy than the original ranging model.

With the technical advancement of transportation electrification and Internet of vehicle, an increasing number of electric vehicles (EVs) and related infrastructures (e.g., service stations with both charging and communication services) are deployed in the intelligent highway systems. Not only can EVs enter the service station areas for charging, but they can also upload/download cached data at service stations to access multiple networking services. However, as EVs are operated individually with their unique travelling patterns, questions arise as how to incent EVs so that both energy and communication resources are optimally allocated. In this paper, we propose an online pricing mechanism of EV charging and data caching for service stations along the highway. First, we design an online reservation system at each EV to decide the best service station to park when the EV enters the highway. Furthermore, based on the variant power system status, an online pricing mechanism is devised to update the charging and caching price based on Q-learning, by which EVs can be motivated to arrive at the designated station for services. Finally, simulation results validate the effectiveness of the proposed scheme in improving the station's utility.

In recent years, with the rapid development of the Internet-of-Things (IoT) technologies, many intelligent sensing applications have emerged, which realize contactless sensing and human-computer interaction (HCI). Handwriting recognition is the communication link between the human and computer. Previous handwriting recognition applications are usually founded on images and sensors, which require significant device overhead and are device dependent. Recently, the revolution of the wireless signal sensing technology has laid the foundation for the intelligent handwriting recognition technology without devices. In this article, we propose WiReader, an adaptive air handwriting recognition system based on wireless signals. WiReader utilizes ubiquitous commercial WiFi devices to process the collected channel state information (CSI), segments the data in combination with activity factors, and then transforms the original signal using the CSI-Ratio model. In order to address the problem of feature extraction caused by handwriting, we utilize the cumulative principal components and multilayer wavelet transform for the transformed signal. Finally, the energy feature matrix is generated and combines with long short-term memory (LSTM) to realize the recognition of different handwriting actions. Extensive real-world experiments show that WiReader achieves an average recognition accuracy of 90.64% leading other applications in three scenarios and has strong robustness to user location, user diversity, and different scenarios.

A reliable recognition of human activities using IoT devices (e.g., on-body wearable sensors) enables various applications, such as fitness tracking, bad habit detecting, healthcare support and elder care support. However, inaccurate results may cause an adverse effect on users or even an unpredictable accident. In order to improve the accuracy in the daily life activities classification, we propose in this paper countable and uncountable activities to better facilitate the understanding of the nature of daily life activities. We design global and local features and their integrated feature set for classifying countable and uncountable activities. The key idea is to examine human daily life activities from different perspectives and attempt to give a comprehensive description of the characteristics of each activity through leveraging the global and local features. By using only one simple accelerometer, our approach is evaluated to be able to recognize daily life activities with higher accuracy than the state of the art, based on one self-collected and another public available dataset.

As the extension of cloud computing and a foundation of IoT, fog computing is experiencing fast prosperity because of its potential to mitigate some troublesome issues, such as network congestion, latency, and local autonomy. However, privacy issues and the subsequent inefficiency are dragging down the performances of fog computing. The majority of existing works hardly consider a reasonable balance between them while suffering from poisoning attacks. To address the aforementioned issues, we propose a novel blockchain-enabled federated learning (FL-Block) scheme to close the gap. FL-Block allows local learning updates of end devices exchanges with a blockchain-based global learning model, which is verified by miners. Built upon this, FL-Block enables the autonomous machine learning without any centralized authority to maintain the global model and coordinates by using a Proof-of-Work consensus mechanism of the blockchain. Furthermore, we analyze the latency performance of FL-Block and further derive the optimal block generation rate by taking communication, consensus delays, and computation cost into consideration. Extensive evaluation results show the superior performances of FL-Block from the aspects of privacy protection, efficiency, and resistance to the poisoning attack.

Application development for the Internet of Things (IoT) poses immense challenges due to the lack of standard development frameworks, tools, and techniques to assist end users in dealing with the complexity of IoT systems during application development. These challenges invoke the use of model-driven development (MDD) along with the representational state transfer (REST) architecture to develop IoT applications, supporting model generation at different abstraction levels while generating software implementation artifacts for heterogeneous platforms and ensuring loose coupling in complex IoT systems. This article proposes an IoT application development framework, named IADev, which uses attribute-driven design and MDD to address the above-mentioned challenges. This framework is composed of two major steps, including iterative architecture development using attribute-driven design and generating models to guide the transformation using MDD. IADev uses attribute-driven design to transform the requirements into a solution architecture by considering the concerns of all involved stakeholders, and then, MDD metamodels are generated to hierarchically transform the design components into the software artifacts. We evaluate IADev for a smart vehicle scenario in an intelligent transportation system to generate an executable implementation code for a real-world system. The case study experiments proclaim that IADev achieves higher satisfaction of the participants for the IoT application development and service orchestration, as compared to conventional approaches. Finally, we propose an architecture that uses IADev with the Siemens IoT cloud platform for service orchestration in industrial IoT.

Software-defined networks (SDNs) offer unique and attractive solutions to solve challenging management issues in Internet of Things (IoT)-based large-scale multi-technological networks. SDN-IoT network collaboration is innovative and attractive but expected to be extremely heterogeneous in future generation IoT systems. For example, multi-technology network, network externality, and nodes heterogeneity in SDN-IoT may seriously affect the flow or application-specific quality-of-service (QoS) requirements. Furthermore, it highly influences security adoption in a network of interconnected IoT nodes. We observe that both QoS and security are interdependent and nonnegligible factors, thus we emphasize that in order to alleviate heterogeneity it is inevitable to study both these factors hand to hand (or vice versa). With this aim, first, we discuss significant and reasonable cases to encourage researchers to study QoS and security integrally in order to alleviate heterogeneity at SDN-IoT control plane. Second, we propose a framework which successfully transforms the m heterogeneous controllers to n homogeneous controller groups. The key metric of our observation and analysis is the SDN controller's response time. Following this, to validate our approach, we use the mathematical model and a proof of concept (PoC) in a virtual SDN ecosystem is demonstrated. From performance evaluation, we observe that the proposed framework significantly alleviates heterogeneity which helps to maintain QoS and enhance security. This fundamental analysis will enable network security individuals to deal heterogeneity, QoS, and security, of SDN-IoT, in more successful and promising ways.

Fog computing is an emergent computing paradigm that extends the cloud paradigm. With the explosive growth of smart devices and mobile users, cloud computing no longer matches the requirements of the Internet of Things (IoT) era. Fog computing is a promising solution to satisfying these new requirements, such as low latency, uninterrupted service, and location awareness. As a typical new computing paradigm and network architecture, fog computing raises new challenges, such as privacy, data management, data analytics, information overload, and participatory sensing. In this article, we present a fog-based hybrid recommender system to address the issue of information overload in fog computing. Our proposed system not only abstracts useful information from the fog environment but can also be considered as an optimization tool due to its ability to provide suggestions to improve system performance. In particular, we demonstrate that the proposed system provides personalized and localized recommendations to users, and also advise the system itself to precache the content to optimize the storage capacity of the fog server.

This article studies multiple unmanned aerial vehicles (multi-UAVs)-enabled wireless-powered Internet of Things (IoT), where a group of UAVs is dispatched as mobile power sources to charge a set of ground IoT devices. Different from the conventional radio-frequency (RF) wireless power transfer (WPT) systems, magnetic resonance-coupled (MRC) WPT systems can guarantee high power transfer efficiency without the complete alignment, which is remarkable. In this article, we extend the charging range by the wired connection between the energy receiving systems and IoT devices. Due to the restriction of carriable energy on the UAVs, designing the shortest possible trajectory for each UAV is necessary. We formulate it as a multidepots multi-UAVs trajectory optimization problem, jointly with constraints of the UAV's energy capacity and the area of the target region, to maximize the resource utilization of UAVs. To tackle this nonconvex problem, we decompose it into two subproblems, i.e., hovering locations selection and multi-UAVs trajectory optimization. For the first subproblem, we propose two approximation algorithms to obtain the near-optimal solution in the sparse networks. Then, we adopt a heuristic algorithm, a memetic algorithm-based variable neighborhood search (MAVNS), to achieve the quasioptimal trajectory rapidly. Finally, extensive numerical results are provided to evaluate the performance of the proposed algorithms. New insights are investigated on the estimation of feasibility that whether the given UAVs with energy capacity constraint can fully charge ground IoT devices within open areas.

The popularization of mobile devices connected to the network promotes the rise and development of the emerging mobile Internet of Things (MIoT). Crowdsensing is a promising mode to perceive data in MIoT, where the collection of sensing data is outsourced to the public crowd carrying mobile devices. However, this crowdsensing mode inevitably makes privacy compromised, due to the workers' sensitive information in the sensing data. As such, how to incentivize workers' participation with privacy preservation becomes a challenge. To tackle this problem, in this article, we propose an auction-based privacy-preserving incentive scheme (APIS) for sensing task allocation in MIoT. Specifically, integrating the idea of software-defined network (SDN), we first present a cloud and edge cooperation-based crowdsensing framework, where the cloud is designed as the controller to collect sensing results from the distributed edge nodes and each edge node outsources sensing tasks to participating workers. To motivate workers' participation, we devise a differential privacy-based auction mechanism, whereby each worker can utilize her privacy budget to control how much privacy can be leaked and decide the sensing precision by the sensing time. Moreover, to maximize the utility of the sensing platform, we design a greed-based algorithm to select the winning workers and determine payments to winners. Finally, we conduct extensive simulations to verify the effectiveness of APIS and demonstrate its superiority.

Federated learning can combine a large number of scattered user groups and train models collaboratively without uploading data sets, so as to avoid the server collecting user sensitive data. However, the model of federated learning will expose the training set information of users, and the uneven amount of data owned by users in multiple users' scenarios will lead to the inefficiency of training. In this article, we propose a privacy-preserving federated learning scheme in fog computing. Acting as a participant, each fog node is enabled to collect Internet-of-Things (IoT) device data and complete the learning task in our scheme. Such design effectively improves the low training efficiency and model accuracy caused by the uneven distribution of data and the large gap of computing power. We enable IoT device data to satisfy ε -differential privacy to resist data attacks and leverage the combination of blinding and Paillier homomorphic encryption against model attacks, which realize the security aggregation of model parameters. In addition, we formally verified our scheme can not only guarantee both data security and model security but completely resist collusion attacks launched by multiple malicious entities. Our experiments based on the Fashion-MNIST data set prove that our scheme is highly efficient in practice.

Compared with individual recommendation, recommending services to a group of users is more complicated because of various users' preference should be considered and introduces new challenging such as fairness, which has never been well studied in current works. In this paper, we propose a novel recommendation scheme called PFGR, which combines a probabilistic model with coalition game strategy, to ensure the accuracy and fairness between groups of users. Given a group of users and a set of services, PFGR models a generative process for service selection in light of several observations: 1) each group is related with several topics; 2) users' decisions on the service selection depends on their expertise, the opinions of members they are familiar with, and group influence; 3) each group contains active users and inactive user, whose activeness contributes to the existence of group. PFGR first estimates the preference of each user on a candidate service via combining user's expertise, inherent connection, and group influence. Then, it determines a group's decision on a service by aggregating the preference of group members using adaptive weights. Finally, PFGR considers users' activeness and employs a strategy based on coalition game to produce a ranked list which is fair to each group member as much as possible. Experimental results on three real-world datasets validate that PFGR can achieve higher Hit Rate and Average Reciprocal Hit Rank than state-of-the-art approaches, which indicates that PFGR attains both the precision and fairness of recommendation.

MapReduce plays a critical role as a leading framework for big data analytics. In this paper, we consider a geo-distributed cloud architecture that provides MapReduce services based on the big data collected from end users all over the world. Existing work handles MapReduce jobs by a traditional computation-centric approach that all input data distributed in multiple clouds are aggregated to a virtual cluster that resides in a single cloud. Its poor efficiency and high cost for big data support motivate us to propose a novel data-centric architecture with three key techniques, namely, cross-cloud virtual cluster, data-centric job placement, and network coding based traffic routing. Our design leads to an optimization framework with the objective of minimizing both computation and transmission cost for running a set of MapReduce jobs in geo-distributed clouds. We further design a parallel algorithm by decomposing the original large-scale problem into several distributively solvable subproblems that are coordinated by a high-level master problem. Finally, we conduct real-world experiments and extensive simulations to show that our proposal significantly outperforms the existing works.

In medical cloud computing, a patient can remotely outsource her medical data to the cloud server. In this case, only authorized doctors are allowed to access the data since the medical data is highly sensitive. Before outsourcing, the data is commonly encrypted, where the corresponding secret key is sent to authorized doctors. However, performing searches on encrypted medical data is difficult without decryption. In this paper, we propose two Secure and Efficient Dynamic Searchable Symmetric Encryption (SEDSSE) schemes over medical cloud data. First, we utilize the secure k-Nearest Neighbor (kNN) and Attribute-Based Encryption (ABE) techniques to construct a dynamic searchable symmetric encryption scheme, which can achieve forward privacy and backward privacy simultaneously. These tow security properties are vital and very challenging in the area of dynamic searchable symmetric encryption. Then, we propose an enhanced scheme to solve the key sharing problem which widely exists in the kNN based searchable encryption scheme. Compared with existing proposals, our schemes are better in terms of storage, search and updating complexity. Extensive experiments demonstrate the efficiency of our schemes on storage overhead, index building, trapdoor generating and query.

Recently, there has been a rapid growth in location-based systems and applications in which users submit their location information to service providers in order to gain access to a service, resource, or reward. We have seen that in these applications, dishonest users have an incentive to cheat on their location. Unfortunately, no effective protection mechanism has been adopted by service providers against these fake location submissions. This is a critical issue that causes severe consequences for these applications. Motivated by this, we propose the Privacy-Aware and Secure Proof Of pRoximiTy (PASPORT) scheme in this article to address the problem. Using PASPORT, users submit a location proof (LP) to service providers to prove that their submitted location is true. PASPORT has a decentralized architecture designed for ad hoc scenarios in which mobile users can act as witnesses and generate LPs for each other. It provides user privacy protection as well as security properties, such as unforgeability and nontransferability of LPs. Furthermore, the PASPORT scheme is resilient to prover-prover collusions and significantly reduces the success probability of Prover-Witness collusion attacks. To further make the proximity checking process private, we propose P-TREAD, a privacy-aware distance bounding protocol and integrate it into PASPORT. To validate our model, we implement a prototype of the proposed scheme on the Android platform. Extensive experiments indicate that the proposed method can efficiently protect location-based applications against fake submissions.

Twitter sentiment analysis has become a hot research topic in recent years. Most of existing solutions to Twitter sentiment analysis basically only consider textual information of Twitter messages, and struggle to perform well when facing short and ambiguous Twitter messages. Recent studies show that sentiment diffusion patterns on Twitter have close relationships with sentiment polarities of Twitter messages. Therefore, in this paper, we focus on how to fuse textual information of Twitter messages and sentiment diffusion patterns to obtain better performance of sentiment analysis on Twitter data. To this end, we first analyze sentiment diffusion by investigating a phenomenon called sentiment reversal, and find some interesting properties of sentiment reversals. Then, we consider the inter-relationships between textual information of Twitter messages and sentiment diffusion patterns, and propose an iterative algorithm called SentiDiff to predict sentiment polarities expressed in Twitter messages. To the best of our knowledge, this work is the first to utilize sentiment diffusion patterns to help improve Twitter sentiment analysis. Extensive experiments on real-world dataset demonstrate that compared with state-of-the-art textual information based sentiment analysis algorithms, our proposed algorithm yields PR-AUC improvements between 5.09 and 8.38 percent on Twitter sentiment classification tasks.

Target-level sentiment classification aims to identify the sentiment of multiple targets in a sentence. Although existing approaches based on neural network have achieved good performance in this task, we find that many approaches tend to give the same predictions for instances that have multiple targets, and this tendency can make a low accuracy for those instances that have different classes for different targets. Based on this observation, we propose MemNetAR, a memory network which can explicitly leverage the adversative relation among multiple targets in a sentence. Specifically, we add an adversative loss to the cross-entropy loss when there are adversative words between targets. The experimental results on public laptop and restaurant datasets prove that our model can improve 0.84% and 0.8% on total test dataset, and improve 2.97% and 2.68% on the dataset consisting of those instances with multiple targets but different classes by leveraging this new adversative information.

With the development of science and technology, mobile phones have gained unprecedented popularity, subsequently the applications based on mobile phone perception have become widespread. Mobile sensing encourages many users to participate in data collection tasks through their mobile phones, but this process raises privacy issues. Previous studies have either used additive homomorphism to protect data privacy or aggregated methods to provide identity privacy protection. However, little research has been done on data dynamics and how to improve aggregation efficiency in multi-user scenarios. To solve this problem, we propose a privacy preserving fog-enabled dynamic data aggregation protocol in mobile phone sensing, named FDDA. In the proposal, we first add fog nodes to our framework, allowing fog nodes to aggregate a set of user data at the same geographical location without identifying the data sources. Then, we design data dynamics strategy to support the user joining and revoking effectively. Finally, we show that our protocol can be well applied in actual scenarios through security analysis and simulation experiments.

Text clustering has been widely used in many Natural Language Processing (NLP) applications such as text summarization and news recommendation. However, most of the current algorithms need to predefine a clustering number, which is difficult to obtain. Moreover, the mutli-label clustering is useful in multiple clustering tasks in many applications, but related works are rarely available. Although several studies have attempted to solve above two problems, there is a need for methods that can solve the two issues simultaneously. Therefore, we propose a new text clustering algorithm called Word2Cluster. Word2Cluster can automatically generate an adaptive number of clusters and support multi-label clustering. To test the performance of Wrod2Cluster, we build a Chinese text dataset, Hotline, according to real world applications. To evaluate the clustering results better, we propose an improved evaluation method based on basic accuracy, precision and recall for multi-label text clustering. Experimental results on a Chinese text dataset (Hotline) and a public English text dataset (Reuters) demonstrate that our algorithm can achieve better F1-measure and runs faster than the state-of- the-art baselines.

We propose to redesign SDN control in IoT lastmiles so as to extend the capability from edge routers to devices (end-node things enabled with SDN capabilities). Our approach put forward existing and new challenges that are impossible to be resolved using the seminal approaches directly. The main challenges we identify are: scalability of sensor nodes/things, maintaining the security of the system, and fulfilling the Quality of Service (QoS) requirement of all IoT applications. Firstly, we elaborate and discuss the aforementioned critical and fundamental challenges that require immediate investigations. Secondly, we propose a policy-driven framework for secure routing and conduct performance modeling and analysis. Further, in the QoS context, we have proposed an intent-based flow offloading scheme to meet the flow-specific QoS requirements. More importantly, we have developed an analysis by modeling TCP-based flows over WiFi, thus forming the required SDN-IoT network, by using mathematics as a tool for reasoning our challenges. With new insights from our analysis, the feasibility of the proposed approach is validated using factors such as path set-up time in SDN-IoT networks, SDN controller/devices throughputs, packets losses and response time of the controller.

Embedding multiple virtual networks (VNs) on a shared substrate network (SN), known as virtual network embedding (VNE), is a challenging problem in cloud platforms. VNE methods can provide strategies to deploy VNs onto SN resources. However, as the scale of VN greatly increases, traditional VNE methods are time-consuming and waste link resource. Meanwhile, traditional VNE methods assign each virtual node of the same VN to different substrate nodes, whereas it is hard to provide larger scale SN to provision the VN. In order to efficiently embed large-scale VNs, multiple virtual nodes from the same VN need to share the same substrate node. We therefore model a repeatable large-scale virtual network embedding (RLSVNE) problem in this study, provisioning large-scale VNs, and propose a heuristic method (Rlsvne) to handle RLS-VNE. Rlsvne pre-processes the VN topology before embedding. In the pre-processing stage, the VN topology is processed through graph coarsening, partitioning, and uncoarsening. After the pre-processing, Rlsvne accomplishes an embedding stage with a topology- aware repeatable embedding solution. 1,000 and 10,000-scale VNE experiments are conducted to demonstrate our Rlsvne. The evaluation results demonstrate that our Rlsvne outperforms three modified heuristics. Rlsvne shows improved performance in reducing substrate cost and fully utilizing substrate resources, achieving high acceptance ratio and revenue values.

User profile release in recommendation systems can apply the user profile perturbation technique to protect user privacy, in which each user sends a perturbed user profile such as the a list of clicked items to receive a recommendation service from a server. The perturbation policy such as the privacy budget determines the recommendation quality and the privacy level, while its optimization usually depends on the known attack model, which is rarely known by the users. In this paper, we propose a reinforcement learning based user profile perturbation scheme that applies differential privacy to protect user privacy for recommendation systems. According to reinforcement learning, the privacy budget to perturb the released user profile depends on the features of the actual user profiles and the released user profiles, and the estimated user privacy level. This scheme enables a user to optimize his or her perturbation policy in terms of both the user privacy level and the received recommendation quality without being aware of the attack model. We evaluate the computational complexity of this scheme and analyze a case study, a privacy aware movie recommendation system. Simulation results show that this scheme improves user privacy protection for a given level of recommendation quality compared with a benchmark profile perturbation scheme.

Federated learning has emerged as a promising solution for big data analytics, which jointly trains a global model across multiple mobile devices. However, participants' sensitive data information may be leaked to an untrusted server through uploaded gradient vectors. To address this problem, we propose a privacy-enhanced federated learning (PEFL) scheme to protect the gradients over an untrusted server. This is mainly enabled by encrypting participants' local gradients with Paillier homomorphic cryptosystem. In order to reduce the computation costs of the cryptosystem, we utilize the distributed selective stochastic gradient descent (DSSGD) method in the local training phase to achieve the distributed encryption. Moreover, the encrypted gradients can be further used for secure sum aggregation at the server side. In this way, the untrusted server can only learn the aggregated statistics for all the participants' updates, while each individual's private information will be well-protected. For the security analysis, we theoretically prove that our scheme is secure under several cryptographic hard problems. Exhaustive experimental results demonstrate that PEFL has low computation costs while reaching high accuracy in the settings of federated learning.

Increasing massive volume of data are generated every single second in this big data era. With big data from multiple sources, adversaries continuously mine private information for potential benefits. Motivated by this, we propose a generative adversarial net (GAN) driven noise generation method under the framework of differential privacy. We add one more perceptron, which is a specifically devised differential privacy identifier. After the generator produces the noise, the discriminator and the proposed identifier game with each other to derive the Nash Equilibrium. Extensive experimental results demonstrate the proposed model meets differential privacy constraints and upgrade data utility simultaneously.

Segment Routing (SR) is an emerging source routing based tunneling technique, which allows source router to steer traffic by encoding segment list in the packet header. Due to its fine-grained control of routing path, SR can be leveraged to facilitate the deployment of Service Function Chains (SFCs). Using SR, multiple segments compose a specific path delivering traffic along a set of ordered Virtual Network Function (VNF) instances. However, when introducing SR into VNF chaining, the segment list depth of SR may face the scalability problem since traffic flows must be steered to traverse a serial of ordered VNFs. To address this problem, we study on segment routing optimization for VNF chaining. Our objective is to minimize the packet overhead of SR for all SFC demands, which indicates the scalability performance of SR. We first formulate the problem as an Integer Linear Programming (ILP) model. Since the ILP model is NP-hard, we then propose a heuristic algorithm named Segment Routing for SFC Steering (SR-SFCS), which is based on the method of backtracking and dynamic programming. Extensive simulation results show that compared with the benchmark algorithms, SR-SFCS can reduce the packet overhead by 23.77% in average.

The 5th generation (5G) wireless network with Unmanned aerial vehicle (UAV) is considered to be one of the most effective solutions for improving the communication coverage. However, UAV is easily affected by the wind, accompanied by a certain time delay during the air communication. Thus the inaccurate beamforming will be performed by the base station (BS), resulting in the unnecessary capacity loss. To address this issue, we propose a novel Recurrent Neural Networks (RNN)-based arrival angle predictor to predict the specific communication location of UAV under the 5G Internet of Things (IoT) networks in this paper. Specifically, a grid-based coordinate system is applied during the data preprocessing to make the training process easier and more effective. Moreover, the RNN model with the highest accuracy can be saved during the training process to ensure the real-time prediction. Simulation results reveal that the RNN-based predictor we proposed is of high prediction accuracy, which is 98% in average. Therefore, a more precise beamforming can be performed by BS to reduce the unnecessary capacity loss, resulting in a more effective and reliable communication system.

In the past few years, it has become increasingly popular to analyze the information obtained to develop services by conducting a decentralized survey of private data for specific populations. Privacy security requirements for data providers force operators to implement reasonable privacy protections. But increasing the investment in privacy protection will also lead to a decline in operator revenue. In this case, operators need to ensure the privacy and security requirements of users while ensuring the sustainability of customized services. To this end, We study the relationship between collecting data quality and operator strategy, quantifying the price of private data, and building a model to maximize operator profitability. Specifically, closed-form solutions for best privacy data prices and subscription fees are designed to maximize the gross profit of service providers. Also includes the collection of data quality factors to ensure that the user perceived quality of service can be guaranteed to a certain extent. Finally, we explored the relationship between spending, subscription fees, and maximum gross profit of carriers during the data collection phase, based on the distribution of different user groups' privacy attitudes. In particular, we also explored the relationship between adding additional noise and collecting data utility in a decentralized privacy protection scenario. The simulation results show that compared with the existing methods, the algorithm can maximize the collected data quality while ensuring the provider's privacy security requirements. In addition, we demonstrate the benefits of our dynamic pricing approach and its applicability to other private data pricing algorithms.

Motivated by cloud servers undertaking heavy encryption of outsourced data for diverse devices, an Economical and High-Quality Encryption Scheme is proposed to alleviate the burden of energy consumption of the servers meanwhile to keep high-quality services. The objective of the scheme is to minimize the cost that combines economy and service quality. For achieving this objective, a two-phase scoring mechanism is proposed. And then based on the above methods and the scoring mechanism, an algorithm achieving the scheme is designed. To evaluate the scheme, some experiments are performed on a heterogeneous platform. The experimental results show that the encryption algorithm can save energy consumption by 47.8% and slightly improve delay rate by 0.93/10000 on average compared with the original one.

Abstract Secure multiparty computation is an important scheme in cryptography and can be applied in various real‐life problems. The first secure multiparty computation problem is the millionaires' problem, and its protocol is an important building block. Because of the less efficiency of public key encryption scheme, most existing solutions based on public key cryptography to this problem are inefficient. Thus, a solution based on the symmetric encryption scheme has been proposed. In this paper, we formally analyse the vulnerability of this solution, and propose a new scheme based on the decisional Diffie‐Hellman assumption. Our solution also uses 0‐encoding and 1‐encoding generated by our modified encoding method to reduce the computation cost. We implement the solution based on symmetric encryption scheme and our protocol. Extensive experiments are conducted to evaluate the efficiency of our solution, and the experimental results show that our solution can be much more efficient and be approximately 8000 times faster than the solution based on symmetric encryption scheme for a 32‐bit input and short‐term security. Moreover, our solution is also more efficient than the state‐of‐the‐art solution without precomputation and can also compare well with the state‐of‐the‐art protocol while the bit length of private inputs is large enough.

Summary In a data sharing group, each user can upload, modify, and access group files and a user is required to generate a new signature for the modified file after modification. There is a situation that two or more users modify the same file at almost the same time, which should be avoided as it gives rise to a signature conflict. However, the existing schemes do not take it into consideration. In this paper, we proposed a new mechanism SeShare for data storing based on blockchain to realize signature uniqueness, which solves the problem of generating signatures for the same file meanwhile by different group users. Specifically, we record every signature of a file in a blockchain in chronological order, and only one user is allowed to add new signature at the end of the blockchain when modification conflicts occur. On the other hand, to provide a secure data sharing service, SeShare introduces an efficient public auditing scheme for file integrity verification when a group user leaves the group. We also prove the security of the proposed scheme and evaluate the performance at the end of this paper. Our experimental results demonstrate the efficiency of public auditing for user leaving.

Summary The high‐bandwidth demands from variety of applications drive a dense wireless local access network (WLAN), which results in a complicated wireless network scene with serious co‐channel interference and energy waste. In this paper, to reveal the interactions between interference and energy efficiency, we propose an interference‐energy efficiency (IFEE) model to quantify the interference impact on the energy efficiency of 802.11 access point (AP) devices. Firstly, we introduce the channel separation and the difference of received signal strength indication (D‐RSSI) as two indicators to extend the classical signal to interference plus noise ratio (SINR) notion and rate adaptive mechanism. Then, these two parameters are integrated into the energy consumption model to establish the IFEE model. Lastly, we conduct extensive measurements with five typical WiFi interference scene in real network to validate the effectiveness of our model. The comparisons between simulation results and real data demonstrate that the proposed IFEE model can quantify the interference and energy efficiency with high accuracy, which can be used for wireless network optimization and protocol design.

There is a growing interest in the use of renewable energy sources to power fog networks in order to mitigate the detrimental effects of conventional energy production. However, renewable energy sources, such as solar and wind, are by nature unstable in their availability and capacity. The dynamics of energy supply hence impose new challenges for network planning and resource management. In this paper, the sustainable performance of a fog node powered by renewable energy sources is studied. We develop a generic analytical model to study the energy sustainability of fog nodes powered by renewable energy sources, by generalizing the leaky bucket model to shape and police traffic source for rate-based congestion control in high-speed fog networks. Based on the closed-form solutions of energy buffer analysis, i.e., the energy depletion probability and mean energy length, we study the energy sustainability in two special but real-happening scenarios. The experimental results show that with proper design the leaky bucket model effectively reflects the energy sustainability of data traffic in fog networks. Numerical results also reveal that the model performance is sensitive to certain traffic source characteristics in fog networks.

An Internet of Things (IoT) system generally contains thousands of heterogeneous devices which often operate in swarms-large, dynamic, and self-organizing networks. Remote attestation is an important cornerstone for the security of these IoT swarms, as it ensures the software integrity of swarm devices and protects them from attacks. However, current attestation schemes suffer from single point of failure verifier. In this paper, we propose an Efficient and Secure Distributed Remote Attestation (ESDRA) scheme for IoT swarms. We present the first many-to-one attestation scheme for device swarms, which reduces the possibility of single point of failure verifier. Moreover, we utilize distributed attestation to verify the integrity of each node and apply accusation mechanism to report the invaded nodes, which makes ESDRA much easier to feedback the certain compromised nodes and reduces the run-time of attestation. We analyze the security of ESDRA and do some simulation experiments to show its practicality and efficiency. Especially, ESDRA can significantly reduce the attestation time and has a better performance in the energy consumption comparing with list-based attestation schemes.

Internet of Things (IoT) can be used to promote many advanced applications by utilizing the sensed data collected from various settings. To reduce the energy consumption of IoT devices, and to extend the lifetime of network, the sensed data are usually compressed before their transmission through compressed sensing theory. By reconstructing the sensed data at the edge of network with more resourceful devices, such as laptops and servers, the intensive computation and energy consumption of the IoT nodes could be effectively offloaded. However, most of the existing data collection schemes are limited in their scalability, because the unified data reconstruction models of them are not suitable for large-scale surveillance scenarios. In our proposed scheme, the whole network is first partitioned into a number of data correlated clusters based on spatial correlation. Then, a data collection tree is built to collect the compressed data in a hybrid mode. Finally, the data reconstruction problem is modelled as a group sparse problem and solved through using an alternating direction method of multiplier-based algorithm. The performance of data communication and reconstruction of the proposed scheme is evaluated through experiments with real data set. The experimental results show that the proposed scheme can indeed lower the amount of data transmission, prolong the network life, and achieve a higher level of accuracy in data collection compared to existing data collection schemes.

In a smart health system, patients' location information is periodically sent to hospitals and this information helps hospitals to provide improved health care services. The location information together with time stamp alone can reveal a patient's private information, such as person's life style, places frequently visited by the person, and personal interests. Thus, it is important to protect the location privacy of a patient. In the existing privacy protection mechanisms, trusted third party (TTP) and location perturbation techniques are used. However, in TTP-based mechanism, an adversary who illegally gets access to TTP server will have access to the private location information. On the other hand, in location perturbation technique, utility of the location information is significantly compromised. In this paper, we propose a location privacy protection mechanism in which location privacy is protected while maintaining the utility of the location data. In the proposed mechanism, a main processing unit attached to a patient's body generates the perturbed location by considering the distance between the patient's location and the preidentified patient's sensitive locations. This adaptive generation of perturbed location, removes the necessity to trust other parties while preserving the privacy and utility of the location data. The validity of the proposed mechanism is demonstrated by simulation results.

Internet of Things (IoT) provides a vast number of devices with heterogeneous characteristics connected to the Internet. As a promising networking paradigm that decouples control plane from data plane, software-defined networking (SDN) is an appropriate architecture for IoT. The SDN paradigm supports deploying traffic flows dynamically by a centralized controller to SDN switches. In particular, the controller configures forwarding rules of SDN switches to steer traffic. However, forwarding rules are usually stored in expensive and power hungry ternary content addressable memory (TCAM), which is very limited in quantity for SDN switches. Thus, the shortage of TCAM becomes a fatal bottleneck for scalable flow management for SDN-based IoT. To this end, we propose a method of forwarding rule multiplexing (FRM) to minimize the total number of forwarding rules in SDN-based IoT. We multiplex different traffic flows traversing through the same path into an aggregated flow with the label of VLAN ID. As a result, multiple forwarding rules could be merged into one multiplexed rule. We also extend the method to SDN protection against link failure, and reduce backup path forwarding rules. We formulate the FRM problem as an integer linear programming model. Since the problem is NP-hard, we design a polynomial algorithm using the Markov approximation technique. Theoretical analysis indicates that the polynomial algorithm generates near-optimal solution. The extensive emulation results show that the proposed Markov approximation-based algorithm reduces the number of forwarding rules by 15.73% in average compared with the benchmark algorithms.

Visual inspection of wheat growth has been a useful tool for understanding and implementing agricultural techniques and a way to accurately predict the growth status of wheat yields for economists and policy decision makers. In this paper, we present a polygonal approach for modeling the growth process of wheat ears. The grain, lemma, and palea of wheat ears are represented as editable polygonal models, which can be re-polygonized to detect collision during the growth process. We then rotate and move the colliding grain to resolve the collision problem. A linear interpolation and a spherical interpolation are developed to simulate the growth of wheat grain, performed in the process of heading and growth of wheat grain. Experimental results show that the method has a good modeling effect and can realize the modeling of wheat ears at different growth stages.

Unmanned Aerial Vehicle (UAV) technology has been widely applied in both military and civilian applications. Recent researches on UAV systems feature in the dramatic augment of the variety and number of equipped sensors, which results in such an issue that multiple UAVs cannot afford to handle the big data generated by a range of sensors in the air. Considering this practical problem, in this paper, we propose a cloud-based UAV system which incorporates the computing capability of the terrestrial cloud into the UAV systems. Relying on proposed cloud-based UAV system, one critical theoretic issue is how to acquire the big data generated by the sensors while guaranteeing a stable operation state of the system. First, we analyze the cloud-based system's on-demand service ability as well as its impact on UAVs' control procedure. Second, the UAV cloud control system is modeled as a network control system. Moreover, the stable condition of the UAV cloud control system is derived, which reveals the relationship between the acquisition rate of sensor data and the stability of the cloud-based UAV system. Finally, simulations are conducted to verify the effectiveness of our theoretical analysis.

Different feedback mechanisms have been developed in large-scale group decision-making (GDM) to provide the decision-makers with advices for preference adjustment with the aim of improving the group consensus level. However, the willingness of the decision-makers to accept these advices is rarely considered in the extant feedback mechanisms. In the field of opinion dynamics, this issue is studied by the bounded confidence model, which shows that the decision-makers only consider the preferences that differ from their own preferences not more than a certain confidence level. Following this idea, this article proposes a large-scale consensus model with a bounded confidence-based feedback mechanism to promote the consensus level among decision-makers with bounded confidences. Specifically, this feedback mechanism classifies the decision-makers into different clusters and provides the corresponding clusters with more acceptable advices based on a bounded confidence-based optimization approach. Finally, through the numerical example and the simulation analysis, the use of the model is introduced, and the effectiveness of the model is justified.

When people express their opinions about a certain issue, they often give uncertain opinions rather than exact opinions. Particularly, these uncertain opinions will evolve in social networks. Therefore, in this paper, we focus on investigating uncertain opinion formation with social networks under bounded confidence. Specifically, we define the uncertain opinions by numerical interval opinions, whose ranges are between zero and one, and the larger width of numerical interval opinions means the more uncertainty of the opinions. Meanwhile, we describe social network structures by ER random graphs with different agents' scales and network connected probabilities. Then, we present the detailed simulation experiments to reveal the strong impact of social network structures on uncertain opinion formation. Simulation results show that: 1) larger agents' scales will yield the smaller ratios of agents expressing the uncertain opinions and larger average widths of uncertain opinions; 2) the average stable time starts increasing and then decreases with the increase in the network connected probabilities; and 3) larger network connected probabilities will yield less opinion clusters and the smaller ratios of the extremely small clusters in all clusters. The obtained results are helpful for the government and public opinion management departments to understand and manage uncertain public opinion evolution effectively.

Today search engines are tightly coupled with social networks, and present users with a double-edged sword: they are able to acquire information interesting to users but are also capable of spreading viruses introduced by hackers. It is challenging to characterize how a search engine spreads viruses, since the search engine serves as a virtual virus pool and creates propagation paths over the underlying network structure. In this paper, we quantitatively analyze virus propagation effects and the stability of the virus propagation process in the presence of a search engine. First, although social networks have a community structure that impedes virus propagation, we find that a search engine generates a propagation wormhole. Second, we propose an epidemic feedback model and quantitatively analyze propagation effects based on a model employing four metrics: infection density, the propagation wormhole effect, the epidemic threshold, and the basic reproduction number. Third, we verify our analyses on four real-world data sets and two simulated data sets. Moreover, we prove that the proposed model has the property of partial stability. Evaluation results show that, compared the cases without a search engine, virus propagation with the search engine has a higher infection density, shorter network diameter, greater propagation velocity, lower epidemic threshold, and larger basic reproduction number.

Social networks are critical in terms of information or malware propagation. However, how to contain the spreading of malware in social networks is still an open and challenging issue. In this paper, we propose a novel defending method through big data based influence modeling. We first establish a social interaction graph based on big data sets of the studied object. Based on the graph, we are able to measure direct influence of individuals by computing each node's strength, which includes the degree of the node and the total number of messages sent by each user to her friends. Then, we design an algorithm to construct influence spreading tree using the breadth first search strategy, and measure indirect influence of individuals by traversing the tree. We identify the top k influential nodes among all the nodes via the social influence strength, and propose an immunization algorithm to defend social networks against various attacks. The extensive experiments show that influence can spread easily in social networks, and the greater the influence of initial spread node is, the more impact it is on the malware propagation in social networks. The proposed method provides an effective solution to the prevention of malware or malicious messages propagation in social networks.

The ever fast growing mobile social video traffic has motivated the urgent requirement of alleviating backbone pressures while ensuring the user-quality experience. Mobile video prefetching previously caches the future accessed videos at the edge, which has become a promising solution for traffic offloading and delay reduction. However, providing high performance prefetching still remains problematic in the presence of high dynamic mobile users' viewing behaviors and consecutive generated video content. Besides, given the fact that making prefetching decision requires viewing history that is sensitive, the increasing privacy issues should also be considered. In this paper, we propose a differential privacy oriented distributed online learning method for mobile social video prefetching (DPDL-SVP). Through a large-scale data analysis based on one of the most popular online social network sites, WeiBo.cn, we reveal that users' viewing behaviors have strong a relation with video preference, content popularity, and social interactions. We then formulate the prefetching problem as an online convex optimization based on these three factors. Furthermore, the problem is divided into two subproblems, and we implement a distributed algorithm separately to solve them with differential privacy. The performance bound of the proposed online algorithms is also theoretically proved. We conduct a series simulation based on real viewing traces to evaluate the performance of DPDL-SVP. Evaluation results show how our proposed algorithms achieve superior performance in terms of the prediction accuracy, delay reduction, and scalability.

Enterprise users at different geographic locations generate large-volume data that is stored at different geographic datacenters. These users may also perform big data analytics on the stored data to identify valuable information in order to make strategic decisions. However, it is well known that performing big data analytics on data in geographical-located datacenters usually is time-consuming and costly. In some delay-sensitive applications, the query result may become useless if answering a query takes too long time. Instead, sometimes users may only be interested in timely approximate rather than exact query results. When such approximate query evaluation is the case, applications must sacrifice timeliness to get more accurate evaluation results or tolerate evaluation result with a guaranteed error bound obtained from analyzing the samples of the data to meet their stringent timeline. In this paper, we study quality-of-service (QoS)-aware data replication and placement for approximate query evaluation of big data analytics in a distributed cloud, where the original (source) data of a query is distributed at different geo-distributed datacenters. We focus on the problems of placing data samples of the source data at some strategic datacenters to meet stringent query delay requirements of users, by exploring a non-trivial trade-off between the cost of query evaluation and the error bound of the evaluation result. We first propose an approximation algorithm with a provable approximation ratio for a single approximate query. We then develop an efficient heuristic algorithm for evaluating a set of approximate queries with the aim to minimize the evaluation cost while meeting the delay requirements of these queries. We finally demonstrate the effectiveness and efficiency of the proposed algorithms through both experimental simulations and implementations in a real test-bed, real datasets are employed. Experimental results show that the proposed algorithms are promising.

The Software-Defined Network (SDN) is a new and promising network architecture. At the same time, SDN will surely become a new target of cyber attackers. In this paper, we point out one critical vulnerability in SDNs, the size of flow table, which is most likely to be attacked. Due to the expensive and power-hungry features of Ternary Content Addressable Memory (TCAM), a flow table usually has a limited size, which can be easily disabled by a flow table overloading attack (a transformed DDoS attack). To provide a security service in SDN, we proposed a QoS-aware mitigation strategy, namely, peer support strategy, which integrates the available idle flow table resource of the whole SDN system to mitigate such an attack on a single switch of the system. We established a practical mathematical model to represent the studied system, and conducted a thorough analysis for the system in various circumstances. Based on our analysis, we found that the proposed strategy can effectively defeat the flow table overloading attacks. Extensive simulations and testbed-based experiments solidly support our claims. Moreover, our work also shed light on the implementation of SDN networks against possible brute-force attacks.

Among various tools invented to help improve people's oral health, water flossers can achieve better performance than traditional and electronic toothbrushes, and are less harmful than dental floss, especially for those with orthodontic teeth or tooth implant surgeries. However, the water flossers available in the market serve no monitoring or recording functions that can help consumers clean their teeth in a more efficient way. To capture users' motions, this study develops a novel smart water flosser, installing an Inertial Measurement Unit (IMU) sensor on the handle of the flosser. We determine the motion cycle using signal processing techniques and extract a set of statistical characteristics from the data set. We then train and compare different machine learning models as classifiers to recognize the motions of the handle. We find that the Random Forest model achieves the best detection accuracy at 97% and 85% of the whole feature set and optimized set, respectively. Finally we implement an Android App that connects the smart water flosser with a Bluetooth module to show the washing area in real-time and record relevant information for further guidance.

The great improvements in both satellite and terrestrial networks have motivated the academic and industrial communities to rethink their integration. As a result, there is an increasing interest on how to combine broadband satellite networks with the clean-slate terrestrial ones, especially with clouds leveraging SDN (Software-Defined Networking) and NFV (Network Functions Virtualization) techniques, for better network openness, flexibility, elasticity and controllability. In this way, customized SFCs (Service Function Chaining) can be deployed at terrestrial and satellite ground segment clouds on demand, significantly reducing OPEX and CAPEX (Operational and Capital Expense). Nevertheless, how to efficiently leverage cloud substrate resources and deploy required SFCs is still challenging, as many issues such as system cost and revenue are involved. Therefore, in this paper, we focus on SFC mappings at SDN/NFV-based terrestrial and satellite ground clouds, and propose a related approach that considers both SF (Service Function) multiplexing and SFC merging, aiming to improve resource utilization efficiency of underlying substrate networks. Extensive simulations are performed and numerical results have verified benefits of the proposed SFC mapping approach.

In data sharing schemes, the file owners should obtain rewards by sharing files with others as they put energy in these files. Therefore, we proposed an incentive data sharing scheme in this paper which encourages users to share data and also supports customization. Customization allows the owners to decide the threshold of access, the importance of each attributive classification which determines users' priority level of file modification and file ownership obtaining when the original owner leaves according to the priority level value. To support a convincing customized data sharing scheme, we introduce the knowledge of blockchain and construct a suitable access structure based on weighted attributes. The blockchain is used to ensure the fairness in incentive. Based on weighted attributes, an attribute set is disposed to a numerical value and the owner of the attribute set is able to obtain the file when the value is not less than the threshold, which is different from the normal access control policy. We prove the security from integrity, privacy and the availability of access key. The performance of the proposed scheme is evaluated at the end of this paper.

The rapid development of big data has brought great convenience to human's lives. The circulation and sharing of information are two main characteristics of the big data era. However, the risk of privacy leakage is also greatly increased when we enjoy the various services of big data. Therefore, how to protect the data privacy in the complex context of big data has become a research hotspot in academic circles. Most of the current researches on privacy protection are divided into two research fields: k-anonymity and differential privacy. Some existing research shows that traditional methods of privacy protection, such as k-anonymity and its extension, cannot achieve absolutely security. The emergence of differential privacy provides a new solution for privacy protection. We draw the lessons from exiting work and propose a new privacy method based on differential privacy: AQ-DP. We propose the first method for classifying quasi-identifiers based on sensitive attributes, which divide quasi-identifiers into associated quasi-identifiers (AQI) and non-associated quasi-identifiers (NAQI). The purpose is not to lose the correlation between quasi-identifiers and sensitive attributes. Our model AQ-DP carries out random shuffling of NAQls., generalizes the AQIs., and adds random noise that satisfies the laplacian distribution to the statistics. We have conducted extensive experiments, confirming that our model can achieve a satisfying privacy level and data utility.

Information-Centric Networking (ICN) is a promising network paradigm for the future Internet. As in the current Internet, selfish routing is also crucial problem in ICN. To the best of our knowledge, however, the selfish routing problem in ICN is remaining an unresolved challenge. To fill this gap, in this paper we propose a Nash Bargaining based content registration (NBREG) method, which is used for register content names (dissemination of content reachability information) from the game theoretic perspective. NBREG allows neighboring domains to cooperate with each other without revealing their internal private information. Based on results from real (inter-domain topology) trace simulations and prototype implementations, we show that neighboring domains can obtain more benefits with NBREG than they register and forward contents selfishly.

With the rich set of embedded sensors installed in smart-phones, an increasing number of applications have been designed based on these mobile sensors rather than on static sensors in urban areas. In Mobile Crowdsourcing (MCS), participant selection is promoted to save energy and entire incentives. Nevertheless, most of the current researches on this problem assume that the system should get the entire information about the participants. As a result, the suitable tasks are always not allocated to the suitable participants. This practice contributes an inaccurate match between a task and participants, which leads to energy and incentives waste. In view of this challenge, we aim to select participants under a more accurate prediction model, rather than assuming that the information of each participant should be obtained in advance. The prediction model is enabled by the big data of participants' historic evaluation, which are used to predict the user action. Furthermore, a greedy method based on an improved singular value decomposition (SVD), named as SVD_G, is proposed to solve this problem. Finally, the proposed SVD_G method is validated by using the large-scale dataset collected from a real-world project (DaZhongDianPing APP).

Recently, there has been an increase in the number of location-based services and applications. It is common for these applications to provide facilities or rewards for users who visit specific venues frequently. This creates the incentive for dishonest users to lie about their location and submit fake check-ins by changing their GPS data. To solve this issue, different distributed location proof schemes have been proposed to generate location proofs for mobile users. However, these schemes have some drawbacks: (1) they are vulnerable to either Prover-Prover or Prover-Witness collusions, (2) the location proof generation process is slow when users adopt a long private key, and (3) their implementation requires some hardware changes on mobile devices. To address these issues, we propose the Secure, Privacy-Aware and collusion Resistant poSition vErification (SPARSE) scheme to generate private location proofs for mobile users. SPARSE has a distributed architecture designed for ad-hoc scenarios in which mobile users generate location proofs for each other. Since we do not integrate any distance bounding protocol into SPARSE, it becomes an easy-to-implement scheme in which the location proof generation process is independent of the length of the users' private key. We provide a comprehensive security analysis and simulation which show that SPARSE provides privacy protection as well as security properties for users including integrity, unforgeability and non-transferability of the location proofs. Moreover, it achieves a highly reliable performance against collusions.

Fast proliferation of mobile devices significantly promotes the development of mobile social networks. Users tend to interact with friends via multiple social networks. Multiple social networks identification is of great significance in terms of both attack and defense. Current methods either focus on the profile matching or network structure to re-identify a specific user. However, the accuracy are not satisfying with relative high error rate. In this paper, we propose a new Friendship learning-Based Identification (FBI) method to discriminate multiple pseudo identities of a real-world individual. We aim at providing potential attack mechanism to following privacy protection research. Firstly, we develop a new identification method based on friendship matching. Then, we implement a weighted mechanism which takes profile, network structure, and friendship into consideration. Furthermore, machine learning is leverage to further optimize the parameters and improve the accuracy. In addition, extensive experimental results show the superior of the FBI comparing to existing ones.

Autonomous Networks has a potential to solve complex and critical management issues in large scale multi- technological networks. Further, the novel paradigms, i.e., Software-Defined Networks (SDN) and Network Function Vir- tualization (NFV) offer unique and attractive solutions for Autonomous Networks or Systems (AS). However, despite of these attractive features, we observed two critical issues in this interlinked multi-technology domain. Firstly, the network externality and nodes heterogeneity seriously effected the flow specific Quality of Service (QoS). Secondly, it influenced se- curity adoption in an network of interconnected nodes. We observed that QoS and security both are non-negligible and inter-dependent factors. This motivates us to investigate solution towards a) alleviating the SDN network heterogeneity at control layer, and b) to strengthen the network security after alleviating the heterogeneity. In this research effort, we have attempted to alleviate the first issue. Firstly, significant and reasonable examples have been cited to motivate researchers to study QoS and security hand-to-hand. Secondly, a theoretical high level frame work has been proposed with the aim to transform the N heterogeneous controllers to n homogeneous controller groups. Following this, we have demonstrated that our approximation method to transform heterogeneous systems to homogeneous groups works well even at high degree of heterogeneity in the network. We have shown our theoretical analysis results using Matlab. Following this, we have shown the Proof of Concept (PoC) of our approach in SDN-NFV ecosystem using Mininet. This early analysis will help researchers to address heterogeneity and security in more effective ways.

In this work, we try to address the problem of memory fragmentation in ternary content addressable memory (TCAM) in programmable data plane (PDP), by designing and implementing a novel network hypervisor for PDP, namely, TPVX. TPVX realizes the virtualization of table resources in PDP with global consideration, i.e., when mapping tenant flow tables to physical switches, TPVX considers their table sizes and the pre-formatted sub-tables in the physical network to improve TCAM utilization and avoid memory fragmentation. Our experimental results verify that with TPVX, the utilization of the table resources in PDP can be improved dramatically and the extra processing latency due to the newly-introduced overheads can be maintained well simultaneously.

In the age of big data, the rapid development of social networking applications has become an improtant data source, while the massive collection of personal data leads to significant privacy concerns. Differential privacy emerged as an effective tool to get access to useful information while provide strong privacy guarantees. However, most the current proposed solutions suppose that all individuals across the network require a uniform level of privacy protection, which rules out of individuals' personalized requirements. Aiming at solving this problem, in this paper, we propose a trust-grained personalized differential privacy mechanism, called TGDP, by combining the notion of trust. Specifically, whenever a user wants to get another user's personal information, the proposed mechanism returns a corresponding private response in which the privacy level selected for each individual depend on the trust value between them in the network. Compared with traditional methods, the scheme can provide a fine-grained differential privacy protection method, while guarantee the utility of social networks. Finally, the scheme is evaluated analytically, and demonstrated experimentally on the real- world data, which reflects its effectiveness and utility.

With the increasing global mobile traffic, there is a trend to deploy network services at mobile edge clouds. Benefiting from the techniques of Network Function Virtualization and Software-Defined Networking, service function chains are enabled to compose a series of required network functions dynamically. As a consequence, most of common-used and IT-based mobile network services can be deployed at MEC cloud networks under the 5G context, remarkably reducing user latency and network traffic. However, as resources in cloud networks are limited, it is challenging to promote the system utilization with guaranteed user experience. Thus, in this paper, we formulate the allocation problem of service functions in MECs as an Semi-Markov Decision Process model and present a value iteration algorithm to find the optimized solution, aiming to increase request acceptance rate. Additionally, we discuss the parameter settings of the proposed scheme under different cases to find higher rewards.

In this paper, we propose a high energy efficiency transmission scheme in multi-beam MIMO satellite systems. Satellite is regarded as a two-way decode-and- forward (DF) relay, where multiple pairs of users exchange information within pair. Zero-forcing transceivers are employed at the satellite. The challenge is that of deriving an accurate yet tractable expression of the system-level energy efficiency (EE) to be used as our objective function. To tackle this challenge, firstly, a closed-form expression of the EE is derived under the assumption of perfect satellite channel. Secondly, based on this analytical expression, we formulate a resource allocation optimization problem for the EE maximization by jointly optimizing satellite power and users power, subject to limited transmit power and minimum quality-of-service (QoS) constraints. Finally, the successive convex approximation technique is invoked to transform the original optimization problem into a concave fractional programming problem, which is then efficiently solved by the existed methods. Simulation results demonstrate the effectiveness of the proposed algorithms.

Due to dramatically increasing information published in social networks, privacy issues have given rise to public concerns. Although the presence of differential privacy provides privacy protection with theoretical foundations, the trade-off between privacy and data utility still demands further improvement. However, most existing works do not consider the impact of the adversary in the measurement of data utility. In this paper, we firstly propose a personalized differential privacy based on social distance. Then, we analyze the maximum data utility when users and adversaries are blind to the strategy sets of each other. We formulize all the payoff functions in the differential privacy sense, which is followed by the establishment of a Static Bayesian Game. The trade-off is calculated by deriving the Bayesian Nash Equilibrium. In addition, the in-place trade-off can maximize the user' data utility if the action sets of the user and the adversary are public while the strategy sets are unrevealed. Our extensive experiments on the real-world dataset prove the proposed model is effective and feasible.

Many datacenters usually process complex jobs such as MapReduce jobs. From a network perspective, most of these jobs trigger multiple parallel data flows, which comprise acoflowgroup semantically. When to schedule the jobs in datacenter or across multiple datacenters, most of current job schedulers have not considered the underlying network traffic load, which is suboptimal for jobs completion times. We present a new deadline-aware coflow scheduling approach called DCS, which takes the underlying network traffic load into consideration while guaranteeing high percentage of coflows that meet their deadlines. DCS aims to alleviate the network congestion in datacenters whose network worload are unbalanced, and it includes two stages for coflow scheduling: Firstly, it generates the task placement proposal by considering the underlying network workload. Secondly, it makes scheduling decision by estimating both task's execution time and transmission waiting time under the previous task placement proposal. The real-world data based simulation results have shown that DCS outperforms all existing solutions on reducing the percentage of coflows that miss their deadlines.

Recently, social network based e-healthcare service has emerged as a promising way to control the spread of infectious diseases. However, the large-scale deployment in reality faces a fundamental challenge to reduce the cost where social features of mobile users and the properties of networks should be considered. To tackle the above problem, this paper presents a green social cyber physical system (CPS) based e-Healthcare scheme to control infectious diseases. Firstly, based on the analysis of social features, the high influential users are selected to inoculate immune drugs when an infectious disease is identified. Secondly, we develop an epidemic spreading model with the dynamic equations to analyze the efficiency of immune strategy. With the proposed model, the spread of infectious diseases can be effectively monitored and the spreading range of the infectious can be predicted. In addition, simulation experiments prove that the proposal can be more efficient to prevent infectious diseases from being spread than conventional methods.

Network Function Virtualization (NFV) is an emerging network architecture, which decouples the software implementation of network functions from the underlying hardware. Telecom operators widely place diverse types of Virtual Network Functions (VNFs) on specified software middlebox. Traffic needs to go through a set of ordered VNFs which forms a Service Function Chain (SFC). However, how to efficiently place VNFs at various network locations while minimizing energy consumption is still an open problem. To this end, we study joint optimization of VNF placement and traffic routing for energy efficiency in telecom networks. We first present the energy model in NFV-enabled telecom networks, and then formulate the studied problem as an Integer Linear Programming (ILP) model. Since the problem is NP-hard, we design a polynomial algorithm using the Markov approximation technique to find the near-optimal result. Extensive simulation results show that our algorithm saves up to 14.84% energy consumption in telecom networks compared with previous VNF placement algorithms.

The demands for high-bandwidth drives a dense wireless local access network (WLAN), which may result in severe co-channel interference and energy consumption increasing. To clearly quantify the effect of interference on energy consumption of 802.11 access devices, it is crucial to measure and model the effect of interference. This paper takes extensive measurements for five different WiFi interference types for downstream UDP transmission in actual environment. Based on experimental measurements, we establish a physical interference-energy efficiency (IFEE) model by reconstructing the signal to interference plus noise ratio (SINR) notion and the modulation and coding scheme (MCS) rate adaptive mechanism to accurately predict the interference impaction. Our experimental measurements demonstrate that interference leads to a decrease in energy efficiency and throughput. Compared with the transmit power, channel separation interference dominates. It is worth noting that the impact of interference with multiple interferers is less than single interferer scene. The simulation experiments verify that our IFEE model can achieve high accuracy of interference and energy efficiency modeling.

Nowadays, cloud computing is increasingly popular. As its important application, outsourcing has aroused great concern. Modular exponentiation is an expensive discrete-logarithm operation and it is difficult for users to calculate locally. Therefore, securely outsourcing modular exponentiation to cloud is a good choice for resource-limited users to reduce computation overhead. In this paper, to outsource modular exponentiation calculation, we dope out a fully verifiable secure outsourcing scheme with single server, so as to eliminate the collusion attacks which occur in algorithms based on two untrusted servers. Meanwhile, our algorithm allows outsourcers to detect any misbehavior with probability 1, which means the checkability of our algorithm shows a significant improvement in comparison to other single server based schemes. Furthermore, to protect data privacy, we propose a new division method to hide the primitive outsourced data. Compared with the state-of-the-art schemes, our secure outsourcing algorithm has an outstanding performance in both efficiency and checkability.

Multimedia big data is one of the cornerstones of Internet of Things (IoT). IoT research subjects naturally connect with multimedia and big data. At the same time, multimedia data and application occupies a large proportion of the landscape of IoT. Moreover, there are many exciting new research directions in multimedia big data based IoT, such as directional sensor networks, video opportunistic transmission, physical object location, image or voice based physical object searching, and so on. Due to the dramatic development of IoT, we have bigger and bigger data sets, and we are stepping into the world of multimedia big data.

We consider the Internet of Things (IoT) with malware diffusion and seek optimal malware detection strategies for preserving the privacy of smart objects in IoT networks and suppressing malware diffusion. To this end, we propose a malware detection infrastructure realized by an intrusion detection system (IDS) with cloud and fog computing to overcome the IDS deployment problem in smart objects due to their limited resources and heterogeneous subnetworks. We then employ a signaling game to disclose interactions between smart objects and the corresponding fog node because of malware uncertainty in smart objects. To minimize privacy leakage of smart objects, we also develop optimal strategies that maximize malware detection probability by theoretically computing the perfect Bayesian equilibrium of the game. Moreover, we analyze the factors influencing the optimal probability of a malicious smart object diffusing malware, and factors influencing the performance of a fog node in determining an infected smart object. Finally, we present a framework to demonstrate a potential and practical application of suppressing malware diffusion in IoT networks.

The crucial role of networking in cloud computing calls for federated management of both computing and networking resources for end-to-end service provisioning. Application of the Service-Oriented Architecture (SOA) in both cloud computing and networking enables a convergence of network and cloud service provisioning. One of the key challenges to high performance converged network-cloud service provisioning lies in composition of network and cloud services with end-to-end performance guarantee. In this paper, we propose a QoS-aware service composition approach to tackling this challenging issue. We first present a system model for network-cloud service composition and formulate the service composition problem as a variant of Multi-Constrained Optimal Path (MCOP) problem. We then propose an approximation algorithm to solve the problem and give theoretical analysis on properties of the algorithm to show its effectiveness and efficiency for QoS-aware network-cloud service composition. Performance of the proposed algorithm is evaluated through extensive experiments and the obtained results indicate that the proposed method achieves better performance in service composition than the best current MCOP approaches.