With the wide use of distributed energy resources, it is important to build efficient and trustworthy coordination among source, grid, load, and storage (SGLS) for the Energy Internet. Blockchain can provide a base of trust, but getting off-chain data through decentralized oracles still faces problems of low efficiency and poor reliability. To address these issues, this paper proposes a four-layer architecture that joins blockchain and oracle services. It also includes a data aggregation and checking algorithm based on threshold signatures and a reputation-based oracle selection method. The main idea of the algorithm is to find reliable nodes more efficiently by using a changing, multi-factor reputation model, and to make the aggregation process faster through preselection and threshold signatures. In this way, it keeps both reliability and efficiency in complex network settings. Simulation results show that the proposed method increases the speed of putting data on the chain and lowers delay, while enhancing the robustness of the oracle network under adverse network conditions. This work providesuseful technical support for building an efficient and dependable distributed-energy coordination infrastructure.

Few-shot knowledge graph completion (few-shot KGC) mines unseen knowledge by leveraging meta-learning and contrastive learning to achieve accurate predictions with limited triples. Recent studies have focused on designing distance or similarity metrics to provide better knowledge representation between entities and relations. However, three issues with negative sampling remain unexplored: 1) the construction of negative queries heavily relies on manual experience in selecting candidate tail entities, 2) the constructed negative queries may mislabel potential true facts, and 3) the varying difficulties of negative queries are ignored. To solve the above issues, in this paper, we introduce curriculum learning into few-shot KGC and propose a novel few-shot KGC framework empowered by an adaptive negative sampling mechanism, which can eliminate the dependence on any additional manual experience, reduce mislabeling, and generate negative queries with appropriate difficulty. Specifically, the proposed framework includes two alternating phases. In the negative sampling phase, we first design a novel positive-unlabeled learning based scoring function with a type-related candidates encoder and then build a variable-speed sliding window based pacing function to select negative queries with appropriate learning difficulty under current training step. In the meta-training phase, we develop an adapted triple-oriented knowledge encoder to provide accurate representation for queries. Experimental results demonstrate that the proposed framework outperforms the state-of-the-art baselines and provides negative queries with appropriate difficulty in few-shot KGC.

Large language models are increasingly deployed across mobile and edge environments, where privacy-sensitive and heterogeneous user data raise critical concerns of copyright infringement, data leakage, and regulatory non-compliance. Ma chine unlearning has thus emerged as an essential capability to remove the influence of specific data without full retraining. However, two key challenges remain open: 1) how to quantify unlearning to enable data valuation without retraining, especially since the massive scale of pretraining makes it infeasible to evaluate the contribution of individual data samples in advance, and 2) how to verify the correctness without retraining to ensure that third-party auditors can efficiently confirm the complete removal of targeted data influence. To address the aforementioned challenges, in this paper, we design a dual-stage machine unlearning framework to quantify the contribution of forgotten data and certify data removal without full retraining, serving as an auditing layer for first-order unlearning methods. Specifically, we design a run-time Shapley value-based unlearned data evaluation mechanism that utilizes a first-order approximation strategy to estimate the marginal contribution of forgotten samples. Moreover, we propose a proof of unlearning mechanism that generates compact, auditable artifacts of the unlearning process to efficiently verify that the targeted data influence has been completely removed. Compared with five state-of-the-art unlearning baselines, our approach achieves effectiveness in data valuation, stronger guarantees of removal correctness, and lower computational overhead.

Multimodal event extraction task aims to identify event types and arguments from visual and textual representations related to events.Due to the high cost of multimedia training data, previous methods mainly focused on weakly alignment of excellent unimodal encoders.However, they ignore the conflict between event understanding and image recognition, resulting in redundant feature perception affecting the understanding of multimodal events.In this paper, we propose a multimodal event extraction strategy with a multi-level redundant feature selection mechanism, which enhances the event understanding ability of multimodal large language models by leveraging knowledge editing techniques, and requires no additional parameter optimization work.Extensive experiments show that our method outperforms the state-ofthe-art (SOTA) baselines on the M2E2 benchmark.Compared with the highest baseline, we achieve a 34% improvement of Precision on event extraction and a 11% improvement of F1 on argument extraction.

Federated Learning has garnered significant attention in practical applications due to its privacy-preserving properties but faces serious threats from backdoor attacks. Current defenses primarily rely on server-side anomaly detection and robust aggregation, but lack systematic strategies for proactively erasing backdoors from the perspective of attackers. To bridge this gap, we propose an efficient and stealthy federated backdoor removal framework. Specifically, our method incorporates a two-stage training approach: reinforced negative learning and positive memory recovery. In addition, we intro-duce a hybrid regularization strategy that combines dynamic L1 regularization with Elastic Weight Consolidation, together with a synchronized differential amplification mechanism for both weights and buffers and a global norm clipping strategy. These components collectively effectively erase backdoor effects, maintain main-task accuracy, and significantly reduce detection risk on the server side. Experimental evaluations demonstrate that, compared to existing backdoor unlearning methods, our approach decreases the success rate of the backdoor attack by up to 11% without compromising the primary precision. Furthermore, our method substantially improves the stealthiness of the update, reducing the L2 norm fluctuations to less than 33% of the baseline levels.

Data heterogeneity and privacy leakage pose inevitable challenges for federated learning in real-world scenarios. These obstacles have motivated the development of numerous personalized federated learning algorithms aimed at acquiring personalized models for individual clients. However, previous work in achieving personalization has typically proposed solutions from the perspective of the entire model or from the level of layer within the network structure. Diverging from them, we introduce a novel perspective, scrutinizing the personalization issue with finer granularity at the parameter level. In this paper, we propose PGPFL, a novel personalized federated learning framework incorporating differential privacy, which rapidly achieves the required local accuracy while providing privacy protection. PGPFL introduces configuring a guard for each parameter of the local network to constrain local updates for personalized learning, and calculate the importance value of each parameter of the current model during the local training process to determine the private parts. Additionally, PGPFL propose a valid client selection mechanism based on the response ratio and local accuracy to decrease single-round training time, enhancing training efficiency. Extensive experiments conducted under various simulated heterogeneous environments demonstrate that, in comparison with existing algorithms, our algorithm improves both average local accuracy and training efficiency.

The development of the Internet of Things (IoT) has led to the rapid growth of the types and number of connected devices and has generated large amounts of complex and diverse traffic data. Traffic identification on edge servers solves the real-time and privacy requirements of IoT management and has attracted much attention, but still faces several problems: (1) traditional machine learning (ML) models rely on artificially constructed features, and the existing deep learning (DL) traffic identification models have reached their performance limit; and (2) insufficient computing resources of edge servers limit the possible improvement in the performance of deep learning models by increasing the number of parameters and structural complexity. To address these issues, we propose a lightweight fusion model. First, the Network-in-Network (NiN) model and Random Forest (RF) model are used on the cloud server to construct a traffic identification fusion model. The excellent representation extraction capability of the NiN compensates for the RF’s dependence on manual feature extraction, and its modular structure is suitable for the subsequent model compression operations. Then, the NiN was distilled. We propose Growth-Adaptive Distillation to lightweight the NiN model, which can reduce the operation of manually adjusting the structure of the student model and ensure the efficiency and low power consumption of the fusion model deployment. In addition, both the RF in the cloud and the distilled NiN are deployed on the edge server. Comparisons with multiple algorithms on two network traffic datasets show that the proposed model achieves state-of-the-art performance while ensuring the use of minimal computational resources.

As the development of B5G and 6G continues to progress, higher network bandwidth and increasingly complex vehicle connectivity are driving greater concurrency in highly dynamic and delay-sensitive transportation tasks within the Internet of Vehicles (IoV). Existing resource allocation methods such as Deep Reinforcement Learning (DRL), Graph Neural Network (GNN), Lyapunov and simple Transformer series often result in insufficient individual consideration or unprioritized attention on key resource characteristics, causing high task execution time cost and energy consumption. To overcome above problems, this paper proposes a Dynamic Self-Feedback (DSF) resource allocation approach. First, DSF models task latency and requirements along with diverse computing power to support allocation and dynamically adjusts the dimensions of self-attention heads according to resource consumption prediction in a self-feedback manner. Then, DSF adjusts dimensions of attention embedding to light and heavy tasks as feedback and leads next round of allocation optimization. Therefore, DSF enables individually tailored and energy-efficient allocation of computing resources for high concurrent IoV tasks. Simulations show the proposed mechanism achieves up to 85% tasks execution efficiency and 38% fewer timeout tasks, more than 50% of low energy consumption tasks after allocation, with almost 100% units having a workload lower than 40%.

With the rapid proliferation of the Internet of Things (IoT), network traffic prediction has become crucial for intelligent network management, enabling more reliable and flexible services for a vast array of IoT devices and applications. The heterogeneous and dynamic nature of IoT networks introduces complex spatial relations and underlying periodic dependencies in spatial-temporal graphs that existing methods struggle to model effectively. In this article, we propose multirepresentation spatial-temporal graph convolutional networks (MRSTGCNs), a novel unified framework specifically designed to address these challenges. MRSTGCN integrates a multirepresentation graph convolutional network (MRGCN) module to model node heterogeneity and complex traffic propagation, and two complementary embedding modules—Historical Embedding and Temporal Embedding—to capture and fuse periodic dependencies across different fine-grained temporal cycles. Extensive experiments are conducted on two network traffic datasets, and the results demonstrate that MRSTGCN achieves state-of-the-art performance with obvious improvements in MAE, RMSE and MAPE on three prediction horizons.

Federated Learning (FL) is a privacy-preserving computing paradigm that enables participants to collaboratively train a global model without exchanging their raw personal data. Due to frequent communication and data heterogeneity of devices with unique local data distributions, FL faces a significant issue with slow convergence speed. To achieve fast convergence, existing methods adjust hyperparameters in FL to reduce the volume of model updates, the number of participating devices, and local iterations. However, most focus on only part of the hyperparameters and primarily rely on analytical optimization. A more integrated and dynamic coordination of all hyperparameters is needed. To address this issue, we first propose an efficient FL framework enabled by rand-m sparsification and stochastic quantization methods. For this framework, we conduct a rigorous theoretical analysis to explore the trade-offs among quantization level, sparsification level, device participation, and local iteration. To improve convergence speed, we also design a Deep Reinforcement Learning (DRL)-based strategy to dynamically coordinate these hyperparameters. Experimental results show that our method can improve convergence speed by at least 8% compared to the existing approaches.

Dataset distillation is utilized to condense large datasets into smaller synthetic counterparts, effectively reducing their size while preserving their crucial characteristics. In Federated Learning (FL) scenarios, where individual devices or servers often lack substantial computational power or storage capacity, the use of dataset distillation becomes particularly advantageous for processing large volumes of data efficiently. Current research in dataset distillation for FL has primarily focused on enhancing accuracy and reducing communication complexity, but it has largely neglected the potential risk of backdoor attacks. To solve this issue, in this paper, we propose three adaptive dataset condensation based backdoor attacks against dataset distillation for FL. Adaptive attacks in dataset distillation for FL dynamically modify triggers during the training process. These triggers, embedded in the synthetic data, are designed to bypass traditional security detection. Moreover, these attacks employ self-adaptive perturbations to effectively respond to variations in the model's parameters. Experimental results show that the proposed adaptive attacks achieve at least 5.87% higher success rates, while maintaining almost the same clean test accuracy, compared to three benchmark methods.

Federated Learning is a promising paradigm, of-fering advantages such as access to extensive datasets and robust data privacy preservation. However, a notable challenge arises from the predominant reliance of most federated learning algorithms on labeled data for model training, leaving a limited presence of algorithms capable of effectively utilizing unlabeled data. In real-world scenarios, individuals using electronic devices inadvertently generate copious volumes of unlabeled data, often accompanied by a small fraction of labeled data. This wealth of unlabeled data harbors valuable information, underscoring the importance of developing high-quality federated semi-supervised learning algorithms adept at harnessing both labeled and unlabeled data. This paper introduces FedIDE, an advanced federated semi-supervised learning algorithm. Our approach integrates the principles of pseudo-labeling and instance discrimination, drawing inspiration from contrastive learning, to unlock the potential of unlabeled data. Concurrently, supervised learning is applied to a labeled dataset to enhance model performance. Additionally, we design the LocalEMA local model update algorithm, which amalgamates local and global models during local training, yielding a hybrid model. FedIDE undergoes extensive testing across multiple datasets, surpassing state-of-the-art baselines.

AI-generated content (AIGC) gained immense popularity across various domains, retrieving valuable training data by using free APIs (Application Programming Interfaces) from various applications and utilizing AI techniques to generate content automatically. However, concerns have been raised regarding unfair payment for the utilization of valuable training data between data owners and AIGC services providers (ASPs). Blockchain oracle can establish trust between them and bridge on-chain and off-chain training data trading. However, the integration of blockchain and AIGC services is challenged by the scalability of oracle consensus. It is essential not only to support a high volume of data requests from ASPs but also to ensure timely and accurate training data responses. To solve the above issues, we first propose an API-based decentralized AIGC data sharing framework and introduce a blockchain oracle to help ASPs retrieve training data from off-chain data owners. We then design flooding-based oracle consensus protocols to achieve scalable and efficient interactions between AIGC and data owners. Theoretical analysis and simulation results demonstrate that the proposed mechanism can significantly reduce communication overheads.

With the advancement of federated crowdsourcing services, the associated privacy concerns have attracted growing attention from both academia and industry. Existing privacy laws impose strict requirements concerning the right to be forgotten for data used in training AI models. In federated crowdsourcing services, the right to be forgotten is guaranteed through federated unlearning. Current federated unlearning solutions encompass a two-step process: first, eliminating model updates associated with the target data to achieve unlearning, followed by retraining among the remaining clients to restore the performance of federated crowdsourcing services. However, this indiscriminate removal of model updates, while safeguarding the privacy of the target data, also greatly undermines the generalization performance of the global model. Moreover, relying on client-side retraining imposes additional economic costs on the federated crowdsourcing service. To tackle the above issues, this paper proposes an efficient federated unlearning framework for federated crowdsourcing services, which is based on adaptive parameter clipping and data-free distillation. We first compute the Fisher information matrix (FIM) to approximate the correlation between the target data and all model parameters, which is utilized to adaptively clip each parameter of the global model. Then, we model the softmax layer of the global model to synthesize pseudo-samples, enabling the retrain process on the crowdsourcing platform for the recovery of generalization performance. We conducted extensive experiments on three datasets, and the results demonstrate that our proposed framework not only possesses outstanding data removal capability but also outperforms the comparison methods in terms of computation time and storage space.

Network traffic prediction is crucial for network operation and management, forming the basis for utilizing big data in decision support. Traditional deep learning methods require extensive, assumed independently and identically distributed(IID) data. However, with IoT development, privacy protection gains importance, resulting in distributed data collection with varying distributions. This leads to a significant performance drop when applying a well-trained model to a new dataset, causing domain shift. To tackle domain shift from inconsistent data distributions and meet privacy protection needs, this paper proposes a spatial-temporal feature learning method within the federated domain generalization framework for network traffic prediction. The ultimately trained model effectively generalizes to an unseen domain, as confirmed by experimental results.

The complexity and dynamism of modern networks pose significant challenges to network management services. Existing technologies often exhibit latency in migrating services after encountering problems. However, adopting a proactive approach through anomaly detection before migration enables the early reservation of resources, thereby ensuring overall performance and stability. This paper proposes a Network Management Service Composition (NMSC) migration method leveraging anomaly detection. The method includes an anomaly detection approach using a Transformer model with a time decay mechanism and a multi-agent reinforcement learning algorithm enhanced by a graph attention autoencoder. First, a time decay mechanism is designed to enhance the self-attention mechanism, allowing the model to capture long-term dependencies while maintaining high sensitivity to recent events. Second, a critic network, enhanced by a graph attention autoencoder, enables the multi-agent algorithm to comprehend the interaction dynamics among agents during computation. Experimental results demonstrate that the proposed anomaly detection algorithm significantly outperforms existing algorithms. Furthermore, the service composition migration algorithm exhibits superior performance in terms of reward value and migration delay, thus proving its effectiveness and feasibility.

With the development of AI-Generated Content (AIGC), data is becoming increasingly important, while the right of data to be forgotten, which is defined in the General Data Protection Regulation (GDPR) and permits data owners to remove information from AIGC models, is also arising. To protect this right in a distributed manner corresponding to federated learning, federated unlearning is employed to eliminate history model updates and unlearn the global model to mitigate data effects from the targeted clients intending to withdraw from training tasks. To diminish centralization failures, the hierarchical federated framework that is distributed and collaborative can be integrated into the unlearning process, wherein each cluster can support multiple AIGC tasks. However, two issues remain unexplored in current federated unlearning solutions: 1) getting remaining clients, those not withdraw from the task, to join the unlearning process, which demands additional resources and notably has fewer benefits than federated learning, particularly in achieving the original performance via alternative unlearning processes and 2) exploring mechanisms for dynamic unlearning in the selection of remaining clients possessing unbalanced data to avoid starting the unlearning from scratch. We initially consider a two-level incentive and unlearning mechanism to address the aforementioned challenges. At the lower level, we utilize evolutionary game theory to model the dynamic participation process, aiming to attract remaining clients to participate in retraining tasks. At the upper level, we integrate deep reinforcement learning into federated unlearning to dynamically select remaining clients to join the unlearning process to mitigate the bias introduced by the unbalanced data distribution among clients. Experimental results demonstrate that the proposed mechanisms outperform comparative methods, enhancing utilities and improving accuracy.

The occurrence of faults increases in edge networks as the service types and component architecture become increasingly complex. Traditional centralized cloud fault prediction technology cannot be directly applied to edge network systems that are typically required to handle real-time data due to their high-computational complexity. Therefore, this article proposes a lightweight fault prediction algorithm for edge networks that realizes fault prediction with cross-layer cooperation. First, for edge devices with limited computation resources, the time feature lightweight extraction method of brain neurology fusion long short term memory (LSTM) based on scene reappearance mechanisms is proposed, which solves the single-step dependency problem of neurons and improves the accuracy. And the LSTM neuron connection method based on pulse dynamics is designed. This method prunes the structure of the LSTM network based on relevant knowledge of biological neurology to realize a lightweight time feature extraction model of fault information. Then, on the edge server side, a spatial feature lightweight extraction method based on a two-way residual structure is proposed. This method uses decomposition convolution to reduce the number of network parameters and realize a lightweight model. Finally, the spatio-temporal correlation features of the extracted fault information are spliced to realize fault prediction. To verify the effectiveness of the proposed model, we compare the improved model with the existing fault prediction model. The experiments show that the algorithm proposed in this article has higher accuracy, lower complexity and lower memory requirements. Therefore, it has high-deployment potential in edge network scenarios.

Object detection through deep learning techniques plays a pivotal role in various industrial applications, such as defect detection. With industries increasingly recognizing the importance of protecting sensitive data, there is a growing interest in collaborative detector training using federated learning (FL). However, existing FL solutions face challenges in effectively addressing object detection tasks with limited labeled data across practical institutions. This challenge is especially pronounced in scenarios with densely packed objects, where obtaining sufficient labels is time consuming and costly. In this article, we present an innovative federated semi-supervised learning (SSL) framework expressly designed for object detection in densely packed scenes (FSSLOD) to overcome above challenges. To achieve this, our approach leverages a teacher-student network on the client side for local SSL and employs a designed consistency loss to align the output of the teacher network with that of the student network. Furthermore, we present an elastic update mechanism to mitigate the intricate issue of data distribution disparity by preventing the inclusion of inadequately trained knowledge into the shared model. Comprehensive evaluations on two real-world object detection data sets demonstrate that the proposed method significantly enhances object detection performance in densely packed scenes while also ensuring data privacy.

AI-Generated Content (AIGC) services are essential in developing the Metaverse, providing various digital content to build shared virtual environments. The services can also offer personalized content with user assistance, making the Metaverse more human-centric. However, user-assisted content creation requires significant communication resources to exchange data and construct trust among unknown Metaverse participants, which challenges the traditional centralized communication paradigm. To address the above challenge, we integrate blockchain with semantic communication to establish decentralized trust among participants, reducing communication overhead and improving trustworthiness for AIGC services in Metaverse. To solve the out-of-distribution issue in data provided by users, we utilize the invariant risk minimization method to extract invariant semantic information across multiple virtual environments. To guarantee trustworthiness of digital contents, we also design a smart contract-based verification mechanism to prevent random outcomes of AIGC services. We utilize semantic information and quality of digital contents provided by the above mechanisms as metrics to develop a Stackelberg game-based content caching mechanism, which can maximize the profits of Metaverse participants. Simulation results show that the proposed semantic extraction and caching mechanism can improve accuracy by almost 15% and utility by 30% compared to other mechanisms.

Network traffic classification is a key technology in network communication management, which is of great significance for building intelligent communication and so on. Due to the difficult and time-consuming process of network traffic labeling, it is difficult to obtain any labeled traffic data in some special networks. At the same time, in a real network environment, there are multiple network traffic domains, and the data distribution of each network traffic domain is different, making it extremely difficult to train a network traffic classification model that performs well on multiple traffic domains simultaneously. Therefore, this paper proposes a network traffic classification method in unsupervised scenarios, aiming to study how to learn traffic knowledge from multiple source traffic domains and achieve accurate classification of unlabeled network traffic without labeled traffic data in the target traffic domain. This paper divides three traffic domains from the data set, namely VPN, nonVPN and nonTor. And three traffic classification tasks of unsupervised multi-source domain are constructed. The accuracy of traffic classification tasks in the source traffic domain is nonTor and nonVPN, and the target traffic domain is VPN reaches 89.76%. The source traffic domain is VPN and nonTor, the accuracy of the classification task is 91.73% when the target traffic domain is nonVPN, and 90.35% when the source traffic domain is VPN and nonVPN, and the target traffic domain is nonTor. Experimental results show the effectiveness of the network traffic classification algorithm proposed in this paper.

Semantic communication constructs a promising and lightweight paradigm for participants to transmit semantic information to each other. However, it suffers from untrust among participants, insecure transmission, and underestimation of the value of semantic information. Blockchain is a decentralized peer-to-peer network that can provide participants with transparent, secure, and trusting environments to implement data sharing. The integration of blockchain and semantic communication is considered a promising paradigm for overcoming the above challenges. However, a unified integration framework has not been studied. Thus, in this article, we first propose a novel blockchain and semantic ecosystems-based framework to share semantic information. We also design the proof of semantic mechanism to solve the garbage-in garbage-out challenge of blockchain. Moreover, we construct a state channel and task-relevant information bottleneck approach-based semantic sharing mechanism to improve the efficiency of semantic sharing. Simulation results show that the proposed mechanisms are verifiable and efficient, and perform better than the compared methods.

Deep convolutional neural networks are hindered from empowering resource-constrained devices due to their demanding computational and storage resources. Structured pruning effectively removes the redundant components from neural networks and obtains compact models. Previous pruning methods usually evaluate the importance of filters from a layer-wise perspective, which is deprived of global guidance. We propose an adaptive pruning algorithm based on relevance scores to evaluate the contribution of each channel by calculating its relevance score from the back-propagation of the neural network's output. Our method identifies and removes channels with low contribution from a global perspective. Unlike previous methods that manually set the pruning rate for each pruning iteration, our method adaptively adjusts the pruning rate. In addition, our method performs satisfactorily with limited data for one-shot pruning in the absence of fine-tuning. The ability to obtain compact models through one-shot pruning with limited data is ideally suited for edge computing scenarios. We validate the effectiveness of our method with multiple combinations of convolutional neural networks and datasets. Our approach outperforms existing pruning methods in scenarios with limited data.

AI-Generated Content (AIGC) services are essential in developing the Metaverse, providing various digital content to build shared virtual environments. AIGC services can offer personalized content with user assistance, making the Metaverse more human-centric. However, it is difficult for participants to exchange data and construct trust among unknown Metaverse participants. To address the above challenge, we propose an integration of blockchain and AIGC to construct decentralized trust among participants. We design a smart contract-based verification mechanism to prevent random outcomes of AIGC services and guarantee the authenticity of digital contents. Given the quality of digital contents provided by the previous mechanisms, we then utilize them as metrics to establish a Stackelberg game-based content caching mechanism to maximize Metaverse participants’ profits. Simulation results show that the proposed caching mechanism can improve utility by 30% compared to other mechanisms.

Federated Learning (FL) is a distributed learning framework that enables data sharing among multiple devices to protect data privacy. Blockchain is a decentralized ledger that can record data securely and reliably. The blockchain-based FL (BFL) framework has been used to share data and computing resources in multiple clusters. However, for the BFL framework among multi-institutional clusters, data sparsity in a cluster is a key issue. Most of the relevant works assume that the data in one cluster is rich enough to build a suitable model, which is not always satisfied in all scenarios. One method to address the problem is that enlarging the size of a BFL cluster that covers as many nodes as possible is one way. However, this method will increase communication overheads and reduce transaction throughput of the blockchain. To solve the above issues, we propose a communication-efficient blockchain-based FL framework called CCFL, which connects multiple BFL clusters to solve the data sparsity issue. We also design a pearson correlation coefficient-based dynamic model filtering mechanism that filters unnecessary models to reduce communication costs and exclude malicious models. Moreover, we illustrate a reliable contribution-based interactive validation reputation mechanism to prevent malicious nodes from participating in the training. We carry out some experiments to show the feasibility and efficiency of the proposed framework.

Blockchain-based Federated learning, called BFL, has attracted widespread attention to construct trust among multiple parties and solve a single point of failure of the central server while protecting privacy. Many researches utilize cluster and cross-chain technologies to improve poor model quality and interoperability between clusters. However, those researches still suffer from 1) high communication overhead when devices of clusters locate far away, and 2) high consensus latency since devices require frequent interactions on consensus. In this paper, we propose a cross-cluster federated learning framework based on state channels, called SCFL, to split devices into multiple clusters according to locations. We also propose a cross-cluster consensus algorithm based on cross-chain and state channels to improve the security and efficiency of off-chain and inter-chain interactions. And we also propose a hierarchical clustering method to make the model adaptable to the partition scenarios where the data is non-IID. Numerical results show that SCFL can effectively solve data sparse problems and improve the system efficiency in non-IID data partitioning cases.

Lightweight models with high accuracy is critical for edge intelligence. Although the Knowledge Distillation (KD) has been successfully applied to reduce the accuracy loss of quantized neural networks, especially for resource-constrained edge devices, the process of pre-training complex high-precision teacher networks in KD however, will bring huge training overhead. Recently proposed online distillation frameworks offer a good solution for teacher-free distillation, but the regularization effect and simple average aggregation of KD further weaken the representation capability of quantized models that have been reconstructed. In this work, we propose Precision-Mixed and Weight-Average Ensemble (PMWAE) consisting of multiple group members and a group leader. PMWAE provides additional knowledge by changing the bit-precision of the activation and generates aggregated weights for each member in group by attention-based mechanism. The ensemble knowledge is further passed to the group leader to obtain the final model. Extensive experiments on the CIFAR-10/100 and ImageNet-1K datasets show that our method outperforms the existing state-of-the-art methods, both on standard convolutions and depth-wise separable convolutions.

Abstract In recent years, Named Data Network (NDN) has become a popular network architecture because of high resource utilization, strong security and high transmission efficiency. Meanwhile, mobile multimedia communication has become the mainstream with the popularization and application of smart terminals. Most of the research on NDN mobility is focused on consumer mobility without taking producer mobility into account. To solve the delay and high cost carried by producer moving, we propose a Double-Lead content search algorithm based on neighbor and proxy and a location prediction algorithm based on traffic features. We use a neural network model to predict a new location of producers and calculate route before switching, which can save the rerouting latency in advance when predicting accurately. In a few cases of inaccurate predictions, we select different search methods according to the distance of the producer’s movement, to complete the Double-Lead search between the producer and the consumer. Experimental results show that DLPNDN can reduce the delay and traffic overhead well in NDN when the producer moves.

The rapid development of digital technology is reshaping the architecture of the Industrial Internet of Things (IIoT). The traditional architecture cannot process vast amounts of data exchanges and provide entities with trust. The future IIoT is expected to be a decentralized architecture in which blockchain and digital twin-driven IIoT can enable trusted data exchanges. However, this architecture cannot obtain huge amounts of external real-time data and isolated data. Moreover, it cannot handle complex industrial computing tasks. Therefore, we combine oracle with decentralized learning to propose a novel IIoT-oriented digital twin architecture. We also propose an effective decentralized collaboration mechanism to support external data and resources exchanges. Moreover, we propose a novel computing collaboration mechanism to expand the learning capabilities of the industrial ecology. Experiments show that our proposed paradigm has less processing time, a more stable process, and better learning ability compared to other paradigms.

Federated learning (FL) lately has shown much promise in improving the shared model and preserving data privacy. However, these existing methods are only of limited utility in the Internet of Things (IoT) scenarios, as they either heavily depend on high-quality labeled data or only perform well under idealized conditions, which typically cannot be found in practical applications. In this article, we propose a novel federated unsupervised learning method for image classification without the use of any ground truth annotations. In IoT scenarios, a big challenge is that decentralized data among multiple clients is normally nonindependent and identically distributed (non-IID), leading to performance degradation. To address this issue, we further propose a dynamic update mechanism that can decide how to update the local model based on weights divergence. Extensive experiments show that our method outperforms all baseline methods by large margins, including +6.67% on CIFAR-10, +5.15% on STL-10, and +8.44% on SVHN in terms of classification accuracy. In particular, we obtain promising results on Mini-ImageNet and COVID-19 data sets and outperform several federated unsupervised learning methods under non-IID settings.

Blockchain-based Federated Learning (FL) technology enables vehicles to make smart decisions, improving vehicular services and enhancing the driving experience through a secure and privacy-preserving manner in Intelligent Transportation Systems (ITS). Many existing works exploit two-layer blockchain-based FL frameworks consisting of a mainchain and subchains for data interactions among intelligent vehicles, which resolve the limited throughput issue of single blockchain-based vehicular networks. However, the existing two-layer frameworks still suffer from a) strong dependency on predetermined and fixed parameters of vehicular blockchains which limit blockchain throughput and reliability; and b) high communication costs incurred by interactions among intelligent vehicles between the mainchain and subchains. To address the above challenges, we first design an adaptive blockchain-enabled FL framework for ITS based on blockchain sharding to facilitate decentralized vehicular data flows among intelligent vehicles. A streamline-based shard transmission mechanism is proposed to ensure communication efficiency almost without compromising the FL accuracy. We further formulate the proposed framework and propose an adaptive sharding mechanism using Deep Reinforcement Learning to automate the selection of parameters of vehicular shards. Numerical results clearly show that the proposed framework and mechanisms achieve adaptive, communication-efficient, credible, and scalable data interactions among intelligent vehicles.

Vehicular ad hoc networks (VANETs) have the characteristics of high mobility, frequently changing topology and uneven distribution, which made it a challenge to design an efficient and robust routing protocol with low latency and high packet delivery rate. Currently, intersection-based routing method and full-path based routing method are two popular solutions for the packet routing in VANETs. Although the intersection-based routing method has better real-time performance, it has the problem of local optimization, making the routing results not global optimal. Although the full-path based method can obtain the global optimal solution, it is weak in dealing with the dynamically changing topological network. Aiming at solving the above problems, this paper designs an intersection-based QoS routing (IQRRL) algorithm, which mainly includes two crutial steps: the next intersection selection and the next hop vehicle selection. In the selection of the next intersection, this paper uses an improved intersection-based routing protocol. In addition to considering connectivity and delay, IQRRL also considers the communication quality from the neighbor’s road to the destination node while evaluating the quality of the neighbor’s road, which minimizes the problem of local optimization. When the next intersection is determined, a road is then determined, and then the next hop vehicle within the road should be chosen to relay the packet forward. In the next-hop vehicle selection step, this paper adopts multi-hop evaluation technology based on reinforcement learning. In addition to using “greedy decision-making” to select the next-hop vehicle, it also comprehensively evaluates whether the next-hop vehicle is still optimal in the future, so that the stability and reliability of data forwarding are improved and local optimal problems are avoided. Besides, this article uses a simulation system to compare IQRRL with other routing algorithms. The result reveals that IQRRL outperforms in terms of packet delivery ratio and transmission delay.

Network traffic classification is a key foundation of traffic management and network security. With the development of traffic encryption technologies and more attention given to user privacy, traditional rule-based and payload-based traffic classification methods have become less effective. To address this problem, recent studies have introduced deep learning-based methods. However, most of these studies do not consider both the flow-level and packet-level characteristics, which we believe are significant in network traffic classification. To further improve the accuracy of traffic classification, this paper proposed DM-HNN, a hybrid neural network based on dual-mode features. First, we treat the packet length sequence as the flow-level feature and the initial byte of the packet as the packet-level feature. Then, we diverge into two paths to analyze the dual-mode features using neural networks. Finally, we combine the two-path features and output the final classification results. We have performed the experiments on public datasets, the results comparing to single-mode and dual-mode traffic classifiers indicate that DM-HNN can achieve excellent performance and has certain effectiveness.

Network function virtualization (NFV) is an emerging technology that uses virtualization technology to provide various services in enterprise networks and reduce costs. However, in cloud edge networks, effective virtual network function (VNF) configuration is particularly difficult, and the system design needs to consider the reliability and energy-saving while meeting the requirements of Quality of Service (QoS). This paper uses the binary integer programming (BIP) model to study the service function chain (SFC) orchestration problem, and designs a federated deep reinforcement learning SFC orchestration algorithm (FDOA). With this method, energy consumption can be reduced and the QoS of users can be improved. In addition, considering the limitations of local deep reinforcement learning (DRL) model training, this paper proposes a federated DRL algorithm to help obtain a more robust model, and simultaneously improve the convergence speed of the model. Among them, we introduce reputation theory during model training to evaluate the reliability of the nodes carrying the DRL model, avoiding the influence of unreliable models on the training effect. Finally, the simulation results show that FDOA has better performance in training time and end-to-end delay compared with other existing algorithms.

Federated learning(FL) is a privacy-preserving dis-tributed learning paradigm in which clients cooperate with each other to train a global model. It is becoming progressively prevalent with the rapid development of edge devices. A critical challenge in federated learning is the data heterogeneity among clients, resulting in the global model generated by standard federated learning being unable to be adapted to all clients. To tackle this problem, we propose the CFedPer for personalized FL, which generates a personalized model for each cluster after clustering to address the deficiency of standard federated learning. Our algorithm is organized into two optimization phases. The pre-start phase clusters clients by our proposed similarity-based clustering model using distribution vector and similarity matrix. In the in-training phase, we represent the neural network as the base layer and personalization layer and propose a novel optimization objective with a regularization term for the personalization layer to achieve a balance between per-sonalization and generalization, preventing over-personalization. Extensive experiments on various datasets and data distributions indicate that the performance of our algorithm is superior to the existing algorithms in terms of average local accuracy and variance among clients.

[J1C2 Presentation Abstract at IEEE SERVICES 2022 for IEEE Transactions on Services Computing DOI 10.1109/TSC.2019.2955486]

Federated learning (FL) is considered to be a promising paradigm to solve data privacy disclosure in large-scale machine learning. To further enhance the privacy protection of federated learning, prior works incorporate the differentially private data perturbation into the federated system. But it is not feasible given the impairment of the model from noise, as adding Gaussian noise to achieve differential privacy (DP) deteriorates the accuracy of the model. In particular, the assumption that the sophisticated system is homogeneous is not realistic for real scenarios. Heterogeneous networks exacerbate noise disruptions. In this paper, we present FedSeC, a novel differential private federated learning (DP-FL) framework which operates with robust convergence and high-accuracy while achieving adequate privacy protection. FedSeC improves upon naive combinations of federated learning and differential privacy approaches with an updates-based optimization of relative-staleness and semi-synchronous approach for fast convergence in heterogeneous networks. Moreover, we propose a valid client selection scheme to trade-off fair resource allocation and discriminatory incentives. Through extensive experimental validation of our method in three different heterogeneities, we show that FedSeC outperforms the previous state-of-the-art method.

Multiaccess edge computing (MEC) provides users with better Quality of Experience (QoE) via offloading tasks to the nearby edge. However, the emergence of new Internet of Things applications with multiple tasks and repeated requests brings redundant computation and transmission to the edge. Meanwhile, the current offloading method based on deep reinforcement learning (DRL) has low sampling efficiency and slow convergence issues for training in a changing environment. Therefore, improving QoE of computation offloading services is still the ultimate challenge. In this article, we devise a collaboration of computing and cache resources among multiple edge nodes, which could reduce redundant computation and transmission. Specifically, we formulate a cache-assisted computation offloading process as a QoE-aware utility maximization problem based on multidimensional indicators. Then, we propose a cache-assisted collaborative task offloading and resource allocation strategy to solve it. This strategy is decomposed into two subproblems. First, to determine and obtain task cache state, we propose a collaborative task caching algorithm, which can improve the hit rate of tasks while balancing network overhead. Second, to acquire offloading and resource allocation decisions efficiently, we propose a metareinforcement learning-based cache-assisted computation offloading method (MCCOM), which can achieve rapid offloading decisions with a few gradient updates and samples. The optimization problem was transformed into multiple Markov decision processes (multiple MDPs). The improved learning process includes metapolicy learning that adapts to multiple Markov decision processes (MDPs) and policy learning for a specific MDP based on metapolicy. Simulation results show that our proposed method outperforms baselines in terms of QoE indicators while achieving rapid convergence and decisions.

In the Industrial Internet of Things (IIoT), model and computing power sharing among devices can improve resource utilization and work efficiency. However, data privacy and security issues hinder the sharing process. Besides, in the process of model sharing, due to the customization of industrial equipment functions and the high separation of model and task types between devices, it is difficult to share model and optimize models among devices with different task requirements. In this article, we propose an adaptively federated multitask learning (AFL) for IIoT devices efficiently model sharing. Inspired by the parameter sharing mechanism, AFL builds a sparse sharing structure by designing an iterative pruning network and generating subnets for each task. Moreover, for better share relevant information, we further propose tailored task mask layers for effectively training specialized subnets, and an adaptive loss function to dynamically adjust the priority between tasks. Extensive experiments show that AFL can successfully fit hundreds of tasks from different devices into one model, which preserves both high accuracy and system scalability, and outperforms other related approaches that naively combine federated learning with multitask learning.

With the rapid development of communication networks, there are more stringent requirements for maintenance management. To carry out smart network maintenance automatically and effectively, this paper uses wearable devices, robots, and Unmanned Aerial Vehicles (UAVs) to collect on-site video data and detect defects in real-time. However, the deep learning models deployed in these devices should have fewer model parameters and less storage space. Therefore, we propose a model compression algorithm based on model pruning and model clustering in smart network maintenance. First, model pruning combines channel pruning and layer pruning and uses deep reinforcement learning to determine the pruning ratio of each layer automatically. This method can effectively compress the width and depth of the model while maintaining the accuracy of the model. Second, although the pruning operation greatly reduces the redundancy of the number of weight parameters, the number of bits of floating-point weights is still redundant. We propose an adaptive model clustering method to cluster the remaining nonzero parameter weights and compress the model. It combines advanced balanced iterative reducing and clustering using hierarchies (BIRCH) clustering and K-meansII clustering and takes the result k value of BIRCH clustering as the input of K-meansII. It can avoid limitations of prior knowledge and reduce clustering time. Simulation results show that the target detection model can reduce parameter redundancy, save storage space, and simplify calculation using the proposed algorithm. In addition, it can be better applied in the smart network maintenance environment.

The continuous development of network technology has driven the emergence of smart devices, and the demand for smart devices interconnection has increased sharply, which requires the identity of devices to be authenticated to carry out secure communication. The traditional certificate-based identity authentication scheme can no longer meet the authentication requirements of massive devices. As an authority that issues and manages certificates, Certificate Authority (CA) creates data islands of intra-domain certificates, increasing the complexity of cross-domain authentication. In order to improve the efficiency of cross-domain authentication, this paper introduces blockchain technology, which can establish trust in an untrusted environment. We propose a multi-CA-based authentication architecture to establish distributed trust and share cross-domain certificate information among multiple domains. On this basis, we design a simplified identity authentication scheme to quickly complete cross-domain identity authentication and reduce authentication overhead. To further improve the efficiency of cross-domain authentication, a cross-domain certificate revocation mechanism is designed. The scheme has passed the formal security analysis, and the simulation results show that the cross-domain authentication scheme is efficient.

With the development of Big Data technology and Internet, the surge of data in the network will cause network congestion and untimely task processing. Additionally, caching content in the core network may cause redundant access of content and backhaul bottlenecks. Due to the increasing requirements of users for task processing efficiency, the centralized maintenance system based on traditional cloud computing cannot meet the current computing requirements. In view of these problems, we propose a content collaborative caching mechanism based on joint decision of download delay and energy consumption. By integrating network coding and content caching technology, the work content maintained in the communication network is deployed near the edge of the network in the form of coding to reduce the redundant transmission of content and acquisition time of content. This article establishes a user QoE satisfaction model, which consists of two indexes that measure time delay and energy consumption. This article proposes a$\varepsilon$-hybrid Q-learning algorithm to optimize the placement of cache files, and made the cache action selection based on the combination of improved heuristic greedy algorithm and simulated annealing algorithm. The experimental results show that the proposed cache strategy can reduce the delay of users downloading content and the energy consumption of content cache, so as to improve the quality of field maintenance work in communication network.

With the development of complex networks and with increasing service demands, service use is becoming more complex and the composition of services is becoming more complicated. In the XaaS (X as a Service) environment, users only care about the QoE of a service and do not care about the composition process of the service. Therefore, it is important to evaluate the reliability of the entire service. In this article, we use Petri Net as a basis for modeling the composition of services. In addition, we consider the problems of shared resources and common cause faults. Both of these problems can cause network congestion and regional failures. We use distance to assess the effects of regional faults and queuing theory to simulate the network congestion process. Moreover, in the simulation, we verify the impacts of regional failures and network congestion on service reliability. We choose the Tree-Based Search algorithm and the Semi-Markov Model as comparison algorithms. The results of our algorithm are related to service time. Our algorithm can timely reflect the impact of regional failure or network congestion, and it can feedback different evaluation results according to environmental changes. Therefore, our algorithm is more comprehensive and has better performance.

Blockchain as a new decentralized chain structure can be used in edge computing to solve the security issue caused by edge nodes in model training. However, large amounts of data exchanges in the process of model training of edge computing reduce the performance of blockchain, and meanwhile, the block needed to be saved in the edge node challenges storage capacity of the edge node. Therefore, in the paper we propose a safe and efficient model training mechanism based on onchain and offchain collaboration. In the mechanism, edge nodes train models locally, store the model parameters in offchain and only return identifiers for model aggregation. By the method, the storage pressure of the edge node is reduced and the efficiency of executing consensus algorithms are increased. Moreover, in the mechanism we design a reputation evaluation model based on confidence factors to avoid the uploading of random and wrong data of edge nodes. Evaluation results show that our schemes can reduce the average delay and resources consumption, increase transaction throughput and maintain security compared with a state-of-the-art scheme.

Federated learning (FL) is a distributed framework for machine learning (ML) model training. Training agents upload local model parameters rather than original training data, and the central server performs parameter aggregation. FL can protect user data privacy and break the information island when training the ML model. Federated Average (FedAvg) is an aggregation method commonly used in the FL training task. The central server calculates the mean value of the local model parameters to obtain the new global parameters. FedAvg assumes that all the training agents are honest, which means the central server lacks terminal agents' knowability. When there are attackers in the training agents, the global model's performance may be deeply affected, and the training task cannot be completed normally. To solve this problem, we propose a Federated Learning method with aggregation based on the Importance of training agent (FedIM), in which the central server performs pre-evaluation on the agent parameters before aggregation, calculates the weights of parameters according to the historical behavior records of training terminals and performs federated aggregation to improve the anti-poisoning ability of learning task. Experiments show that our method can effectively improve the global model's anti-poisoning ability and accelerate the training speed compared with the FedAvg method when malicious agents are involved.

The blockchain system allows various trans-actions and information storage to be executed in a decentralized manner, while smart contracts require multiple nodes to be executed in the local sandbox environment according to preset settings to ensure the consistency of each node, which makes smart contracts unable to proactively obtain data from the outside world. Decentralized oracle can realize the acquisition of off-chain data with a low speed under the premise of ensuring the decentralization of the blockchain. Some oracles use on-chain data storage and maintenance to speed up data acquisition, but this will face higher costs of data storage and maintenance, so current oracles cannot simultaneously ensure privacy and security while taking into account execution cost and processing speed. In this article, we propose Select-Storage, a new oracle design pattern to achieve low operating cost and high processing speed without compromising security. Through experimental analysis, and comparison with other design patterns in processing time and on-chain and off-chain call costs, we have proved the superiority of the Select-Storage design pattern.

The traditional centralized data processing model represented by cloud computing cannot meet the data processing requirements that are gradually tending to the edge. Therefore, a new distributed computing model coordinated by the end devices, edges and cloud has become the main development direction. However, artificial intelligence algorithms that are widely used in cloud-only approach are difficult to embed in resource-constrained distributed frameworks. To address this issue, we propose Triple-partition Network, a neural network model augment with three exit points. The structure of three exit points allows to segment the traditional neural network and deploying them on the end devices, edges, and cloud. By setting up suitable exit points through the Entropy Topsis comprehensive evaluation model, part of the data can exit the network in advance to improve the efficiency of computing services. In this experiment, the classic neural networks (Alexnet, Resnet) are used to study the Triple-partition Network on a state-of-art platform and show that trained Triple-partition Network can greatly reduce the end-to-end latency by over 3x while achieving high accuracy.

The power system is an infrastructure for industrial manufacturing, and its availability is relevant to industrial systems. The smart grid combines communication systems with sensing devices to provide intelligent management tools for fault monitoring and processing of the power grid. Regional failures caused by natural disasters can have a large impact on the power system. To reduce the impact of disasters, an efficient network repair strategy is needed. This article focuses on the emergency repair strategy of a power communication network under disaster conditions and seeks to ensure the operation of the power system with fewer repairs. Combined with the analysis of cascading failures and regional failures, a communication network fast repair algorithm is proposed. The coupled network is constructed in the simulation part to verify the algorithm. The results show that with a limited number of node repairs, the algorithm can ensure the highest percentage of workable nodes.

With the development of Industry 4.0, an increasing number of industrial Internet of Things (IIoT) mobile devices (MD), which constantly transmit data at any time, are working on the production line. However, due to node movement, signal attenuation, or physical obstacles, data must rely on the transmission of relay nodes to finally reach the destination node. Based on this scenario, in this article, we propose a cooperative link prediction model (CLPM) using a stacked denoising autoencoder (SDAE) to predict links of the IIoT-based MDs at the next moment through historical link information. The layer structure of the SDAE model is partitioned so that the local MD and edge servers can cooperatively process the link prediction tasks. Experimental results show that our proposed CLPM outperforms others in terms of prediction performance and execution delay.

With the development of information technology, the network consists of various proprietary hardware devices, and the use of these devices brings problems. To solve problems, network function virtualization is proposed, which decouples the software and hardware in the network, and deploys the existing network function devices to a common physical platform. However, network virtualization needs will inevitably face reliability problems during resource virtualization and service function chain deployment. This article proposes a service function chain reliability evaluation method and reliability optimization algorithm. The composition relationship and reliability influencing factors of service function chain were analyzed, including resource preemption, common cause failure, fault recovery and redundant backup. The service function chain was modeled as a Petri net model, and reliability evaluation results related to execution time were obtained. Based on the reliability assessment results, a VNF migration strategy is designed, with reliability as the optimization goal while considering costs. Simulation results show that, compared with the reliability optimization strategy based on backup, our algorithm costs less and reduces the impact of resource preemption on service reliability.

As things currently stand, the blockchain industry is siloed among many different platforms and protocols resulting in various islands of blockchains. Restrictions regarding assets transfers and data migration between different blockchains reduce the usability and comfort of users, and hinder novel developments within the blockchain ecosystem. Interoperability will be the main topics of next-generation blockchain technologies. In this paper, we focus on how to enable interoperability between two heterogeneous blockchains in the context of data migration. We first build an cross-chain data migration architecture based on data migration oracle. Second, we design a data migration mechanism based on former architecture. By employing the proposed data migration architecture, it is equivalent to opening a secure channel between two heterogeneous blockchains allowing secure data migration. By applying data migration mechanism, the confidentiality, integrity and security of migrated data can be well guaranteed.

Content pre-caching is a kind of significant technology to lower response delay and improve network performance, especially for the delay-sensitive services in dynamic vehicular edge networks. Therefore, in this paper, we propose a zone-based content pre-caching strategy, which aims to implement an active content caching through two algorithms: pre-caching zone selecting algorithm- PCZS and pre-caching node selecting algorithm- PCNS. Firstly, we organize the edge servers (ESs) with a zone-based way at the edge, and assign a Manager node to collect the information of each zone; with the help of road topology and tables information recorded in Manager nodes, PCZS can predict the vehicle motion and zone sojourn time with a high accuracy, and further get a content pre-caching zone by comparing estimated request delay and zone sojourn time; then, PCNS checks whether the content has been cached in the CST of the zone selected by PCZS, if CST hits, the pre-caching process is over, otherwise, by combining ES centrality, load degree with content popularity, we perform PCNS to select a specific ES node to pre-cache the content; simulation results show that our strategy has a higher prediction accuracy and dynamic adaptability, it also outperforms in terms of average response delay and cache hit ratio.

The introduction of edge computing (EC) in intelligent driving allows the vehicle to offload tasks to the EC server closer to the vehicle side, creating a new paradigm for task offloading and resource allocation. The movement of the vehicle, the time sensitivity of the processing data, and the resource allocation of the EC server have become bottlenecks of the rapid development of intelligent driving. In this article, we jointly considered the problems of the network economy and resource allocation. In order to eliminate dependence on third parties, we propose a resource transaction architecture based on the blockchain. Moreover, we propose the dynamic allocation algorithm of edge resources (DAERs) based on the double auction mechanism to maximize the satisfaction of users and service providers of edge computing (SPs), where the DAER algorithm is implemented in the form of smart contracts in the blockchain architecture. In particular, we propose the state search algorithm that can improve the prediction accuracy of the staged destination of the vehicle to help allocate resources reasonably. Through simulation experiments, we verify the superior performance of the DAER algorithm in terms of resource utilization rate and the satisfaction of both parties participating in the auction.

To alleviate the security concerns caused by the openness of the edge computing network and meet the time-sensitive requirements of the edge devices' collaborative tasks, an effective trust evaluation mechanism is needed urgently to resist multi-attacks from various malicious devices. In this work, a light- weight trust mechanism based on multi-source feedback is proposed for edge computing. First, we design a light-weight data-processing algorithm executed in edge brokers and edge devices, which could reduce the data transmission pressure in communication networks effectively and work efficiently in large-scale edge networks. Then, a comprehensive evaluation method is designed for edge brokers based on the Dempster Shafer theory and multi-source feedback mechanism, which makes our mechanism more reliable and pluralistic when resisting various multi-attacks at the same time. At last, we originally develop a neural network in the centralized cloud to update edge brokers' hyper-parameters and weights of the key factors by auditing trust evaluation results uploaded from the edge network according to deep Q-learning algorithm, which are usually weighted manually and subjectively in traditional schemes. The experimental results show the proposed trust mechanism outperforms existing methods in reliability and calculation efficiency when resisting various malicious attacks.

With the development of the edge computing and cloud computing technology, the cloud-edge collaboration framework is proposed as a new effective computing architecture and applied in many fields. However, due to the openness of the edge networks, the security of cloud-edge framework is an unavoidable problem and most recent trust mechanism could not resist mixed malicious attacks at the same time. In this work, a light-weight and reliable trust mechanism based on the improved LightGBM algorithm is originally proposed to evaluate the credibility of edge devices. First, we design a light-weight trust mechanism for edge devices to process raw interaction data and extract trust features, which reduces the amount of data transmission and the pressure on the communication networks. In addition, an evaluation algorithm based on the entropy weight method (EWM) and punishment factors is designed for edge brokers to distinguish the malicious devices from the normal ones, which performs great against mixed malicious attacks. At last, we propose an improved LightGBM algorithm developed in the centralized cloud to learn other researchers' evaluation methods and check the evaluation uploaded from edge brokers, which could make the punishment factors of edge networks weighted adaptively with the change of edge networks. The experimental results show the proposed trust mechanism outperforms existing methods in the accuracy and discriminating speed under mixed malicious attacks.

Task offloading has been a hot topic in the field of edge computing. Resources fairness of edge computing servers which is the destination of task offloading directly impacts life of server and the process quality of task. In this paper, we propose a subtask-virtual machine mapping model (subtask-VM mapping model) to complete task offloading from the terminals to the servers. Considering the reasonable allocation of server resources, we also propose stack-based cache mechanism (SCM) to ensure the fairness of server resources allocation. We transform the problem of mapping model solution into the problem of optimal matching in the bipartite graph, and verify the performance of our algorithm by contrast experiment. In particular, the fair performance of our algorithm for server-side is over 84%.

Various problems arise in the maintenance of communication networks. For example, on-site maintenance personnel have insufficient work experience. Devices used for maintenance work have limited computing resources and battery life. Moreover, most maintenance systems still use the centralized single processing mode of traditional cloud computing, which increases the data center computing pressure and slows the data flow. To overcome these problems, we propose a communication network edge maintenance system based on smart wearable technology and introduce computation offloading technology for mobile edge computing (MEC). Before offloading, we propose a multimerged computing sorting segmentation (MCSS) algorithm to divide a part of the task to offload. When making an offloading decision, we access a suitable MEC service node for each user with the lowest transmission cost and establish a related model. We use an improved Kuhn-Munkras (KM) algorithm that considers fairness among users to solve this model. After that, we propose a dynamic energy-efficiency awareness strategy. When tasks are processed locally, we optimize the CPU clock frequency. When tasks are offloaded, we adaptively allocate the transmission power. Finally, we conduct a simulation experiment. The results demonstrate that the proposed scheme can reduce the transmission cost and improve the performance, thereby increasing the level of on-site maintenance work.

To solve the issue of scarce spectrum resources of the existing cellular network, LTE-U that expands LTE service to the unlicensed 5GHz spectrum is proposed. However, the centralized medium access control protocol of LTE largely decreases the performance of Wi-Fi networks operating in the same unlicensed spectrum. In the paper, to solve the problem, we propose a new mechanism based on the duty-cycle method. It can adaptively adjust the percentage of the airtime used by a LTE Small cell Base Station (SBS) according to the bandwidth of the licensed spectrum of the SBS and downlink data rate demands of the SBS users to maximize throughput of the SBS network on the unlicensed spectrum while ensuring fairness between the Wi-Fi and SBS network. The fairness is based on 3GPP proposed fairness coexistence criterion. To ensure the fairness, we propose a method to construct a W-Fi network offering the same level of the SBS traffic load, and throughput maximization of the SBS is formulated as a constrained non-linear optimization problem solved by an optimal algorithm. We evaluate the proposed mechanism from two aspects. The first is to prove the proposed Wi-Fi network construction method is valid. The second is to evaluate the performance of our proposed method. Simulation results show that our approach is valid and it can maximize the throughput of the SBS network and ensure the fairness criterion.

Mobile edge computing is a promising paradigm to compensate for the lack of traditional cloud computing, which has a variety of application scenarios. However, the migration of user-generated data in edge networks is a key issue which involves in transmission costs, the mobility of users, transmission resources, etc. In this paper, we focus on migrating deadline-varying user-generated data to edge servers, considering the tasks characteristics and contact patterns between nodes. We design a heuristic algorithm and propose the online algorithm using real-time information to save the cost of transmission. Further, we conduct the extensive simulations to demonstrate the effectiveness of our algorithms.

Opportunistic offloading can be utilized to offload computing tasks and traffic data in Mobile Edge Computing (MEC). To improve the ratio of successful data offloading and reduce unnecessary data redundancy in opportunistic forwarding process, some methods of evaluating a device’s forwarding capability are proposed. However, most of these methods do not consider the temporal impact from device mobility and the efficiency influence from the capability computation process. To settle these problems, we proposed a Transient‐cluster‐based Capability Evaluation Method (TCEM) to evaluate a device’s data forwarding capability. The TCEM can be divided into two steps. The first step aims to reduce computational complexity by evaluating a device’s possibility of contacting the destination within a time constraint based on the transient cluster generated by our proposed Transient Cluster Detection Method (TCDM). The second step is to calculate a device’s probability of directly and indirectly forwarding data to the destination. The probability as a metric of evaluating a device’s forwarding capability can be used in different data forwarding strategies. Simulation results demonstrate that the TCEM‐based data forwarding strategy outperforms other data forwarding strategies from the aspect of the proportion of the data delivery ratio to the data redundancy.

Fault-tolerant topology control is a critical problem in WSNs. It is important for improving network lifetime and reliability. In this paper, we present a novel algorithm FTMRT, which ensures Fault Tolerance by constructing a Multi-Routing Tree. We firstly construct a multi-routing tree of the initial topology, which ensures there are at least k-disjoint paths from each sensor to the set of supernodes. And then each sensor adjusts its transmission power according to the multi-routing tree to form the fault-tolerant network topology. In the topology maintenance phase, topology reconstruction is invoked each time there are some node fail and the supernode connectivity is broken. The effectiveness of the proposed algorithm is validated through simulation experiments.

The wireless sensors deployed at the highway can ensure the safety of the traveling vehicles. However, the ribbon deployed wireless sensor network in the roadside infrastructure can easily to generate energy hole. Cooperative communication between sensors and vehicles is an effective way to improve this situation and reduce the energy consumption of the sensors. A cooperative relay selection algorithm based on residence time (CRSR) and cooperative relay selection algorithms based on prediction (CRSP) are proposed in this paper. In order to improve the data transfer amount of the vehicle, residence time is considered in CRSR when sensors select the vehicles. To further improve energy efficiency CRSP considers arrival time of the vehicle to store collected data in the delay tolerance situation. Energy is also considered in CRSP to reduce the energy consumption and ease energy hole. Simulation results show that the CRSR and CRSP methods can reduce the energy consumption and prolong sensor network lifetime than the traditional algorithm.

Object Tracking Sensor Network (OTSN) is considered one of the most energy consuming applications of wireless sensor network. OTSN is used to track moving objects and report their newest location which consumes a large amount of energy. However, energy of sensor node is limited and the movement of objects generally follows some definite patterns. We can reduce the energy consuming by predicting the next location of an object to keep irrelevant sensor nodes sleepy as long as possible. In this paper, we propose an energy-efficient prediction-based tracking algorithm called Improved Mining Pattern (IMP). This algorithm predicts the next active sensor node based on the backward dependence. The predicted paths can be updated partly fast through clustering. Besides, IMP reduces the long distance communication between sensor nodes and the base station. In addition, missing objects can be tracked again quickly through recovery algorithm which is based on prediction results. Moreover, this algorithm can track multi-species simultaneously. Experimental results show that IMP behaves better than other algorithms in reducing the energy consumption and the missing rate.

MapReduce is an important distributed programming model for large-scale data-parallel applications like web indexing, data mining, and scientific simulation. Hadoop is an open-source implementation of MapReduce and it is often applied to short jobs for which low response time is critical. When the cluster nodes are homogeneous, Hadoop has a good performance. In practice, the homogeneity assumptions do not always hold. In heterogeneous environment, there are various devices which vary greatly in the capacities of computation, communication, architectures, memories and power. When different nodes process the same amount of data, load balancing problem occurs. In this paper we address the problem of how to assign data after Map phase to balance the execution time of each Reduce task by proposing a novel load balancing algorithm based on nodes performance (LBNP), in which the input data of poor performance nodes are decreased. Simulation results indicate that all the Reduce tasks can be completed in the same time which shortens the whole Reduce phase. Thus the efficiency of MapReduce is improved.

In the management of WSNs, the mechanism of fault detection and location based on active probing has been widely applied. The main optimal direction of active probing is to maximize the coverage of nodes in the network by sending minimal set of probes from probe stations. Therefore, before probing, selecting optimizational station set to improve reachable rates of probed nodes has significant influence on detection effect of active probing. In this paper, we propose an optimizational strategy of probe station selection (PSS) by Genetic Algorithm (GA) and achieve the improvement of confirmed achievable rates of probed nodes. Meanwhile, lower runtime cost and more reasonable usage of energy are reached.

The main role of wireless sensor networks is to collect environmental data. As the sensor nodes are vulnerable and work in unpredictable environments, sensors are possible to fail and return unexpected response. Therefore, fault detection and recovery are important in wireless sensor networks. In this paper, we propose a fault detection algorithm based on support vector regression, which predicts the measurements of sensor nodes by using historical data. Credit levels of sensor nodes will be determined by a contrast between predictions and actual measured values. In this paper we also propose a fault recovery algorithm according to the node credit levels combined with genetic algorithm. The simulation results demonstrate that the algorithms we propose work well in failure detection rate, fault recovery speed and energy consumption.

Since wireless sensor network has limited resources, it's important to design its task allocation algorithm reasonably to reduce energy consumption. The contract net is simple and flexible so that it can meet the needs of the wireless sensor network. In this paper, we introduce the improved C-MEANS algorithm to cluster nodes to decrease the number of bidders, and at the same time, the LMS algorithm is adopted to predict the bid value of the nodes. The simulation results show that the energy consumption and traffic flow are reduced, and the bid value more accurately reflects the status of the node when allocated tasks, which increased the complete rate of network tasks.

Coverage control is one of the most important technologies in Wireless Sensor Network (WSN). In the precondition of better coverage quality, how to format optimal coverage with least sensors is a significant problem to be solved. A new incomplete coverage control based on target tracking sensor network which called mobile-constrained optimal target tracking coverage algorithm (MCOTT) is presented. In our approach, static sensors will be pre-deployed, collaborating with mobile sensors to achieve an optimal coverage which based on a target trajectory prediction model. Simulation results show that, MCOTT has more advantages like good robustness, high level of target coverage, low energy consumption. The algorithm can save the number of sensors and prolong the network lifetime effectively.

In WSN, Clustering Routing Algorithm can effectively reduce network energy consumption and prolong network lifetime well. But existed Clustering Routing Algorithms are usually location-based, where data correlations are not considered. There is still data redundancy in the terminal. This paper proposes a data correlation-based virtual clustering approach. It integrates the advantages of clustering technique and data correlation. Nodes that are good data correlated will be partitioned in the same virtual cluster. The experimental results show that the proposed algorithm can reduce the amount of messages sent by the nodes, and reduce the energy consumption. The network lifetime is prolonged as well.

In this paper, we present a distributed fault detection algorithm based on k-means clustering for WSN. The nodes within a cluster are divided into three sub-clustering according to their measurements' similarity. We conclude the sensor nodes' working state from the N recent states of sub-clustering, so as to detect, locate, and get rid of the fault nodes. Simulation results show that the k-means cluster fault detection algorithm has a better performance than the distributed Bayesian algorithms. Moreover, the computational complexity of the proposed algorithm is low.

With the development of high-speed networks, the challenge of effectively analyzing the massive data source for anomaly detection and diagnosis is yet to be resolved. This paper proposes a new flow-based anomaly detection method based on summary data structures and combinations of traffic features. Using IPFIX flow records as input, parallel sketches are established for chosen traffic features respectively. For each sketch, we use Holt-Winters forecasting technique to achieve their forecast sketches and deviation matrixes. When the deviation exceeds a certain threshold, sub-alarms will be generated. According to the characteristics of various attacks and combinations of traffic features, sub-alarms can be merged into final alarms. While sketches of flows are being constructed, destination addresses are recorded in linked lists which are used to locate victims by a series of set operations. This method can not only detect the existence of anomalies in near real time, but can roughly indicate the anomaly types and locate abnormal addresses.

This paper studies task allocation in cluster-based Wireless Sensor Network (WSN) using negotiation model. We study both the negotiation reasoning model and the negotiation protocol for task allocation to achieve energy efficiency while balancing nodes energy. Reasoning model determines the offer generate scheme and gives control over negotiation process. A time depending Boulware function is used as the concession strategy in reasoning model to balancing efficiency and utility. Contract net protocol is used as negotiation protocol to regulate the interaction style of nodes. The goals of this study are: 1) energy efficiency task allocation; 2) maintaining energy balance of nodes in WSN after the task to prolong the network life cycle. Experimental results using this cluster-based negotiation model task allocation approach verify its performance.

Traffic matrices are very essential for many network engineering tasks: for instance, load balancing, capacity planning, routing protocol configuration. However, measuring these traffic matrices directly is difficult and costly. Hence many methods have been proposed to estimate these traffic matrices based on link load measurements and other more easily available data. This paper presents an iterative algorithm to estimate traffic matrix without differentiating the access links and the peering links and our algorithm get a similar performance with the Minimal Mutual Information method which requires that difference. Experiments on real backbone network data have also demonstrated that our algorithm is accurate and robust to measurement noise.

Mobility Model for mobile users is an important part of simulation of the wireless mobile network's resource management and optimization. The Manhattan Mobility Model proposed by ETSI is highly versatile. However, its random direction selecting strategy lacks of objectives, which means the scene of hotspot attracting user mobility could not be simulated accurately. In order to simulate the user mobility attracted by hotspots, a novel Hotspot Attraction Driven User Mobility Model (HADUMM) is proposed based on the Manhattan Mobility Model, and a Direction Deciding Algorithm (DDA) is designed. The HADUMM's realistic significance is evaluated.