Automated Uncrewed Aerial Vehicle (UAV) landing is crucial for autonomous UAV services such as monitoring, surveying, and package delivery. It involves detecting landing targets, perceiving obstacles, planning collision-free paths, and controlling UAV movements for safe landing. Failures can lead to significant losses, necessitating rigorous simulation-based testing for safety. Traditional offline testing methods, limited to static environments and predefined trajectories, may miss violation cases caused by dynamic objects like people and animals. Conversely, online testing methods require extensive training time, which is impractical with limited budgets. To address these issues, we introduce GARL, a framework combining a genetic algorithm (GA) and reinforcement learning (RL) for efficient generation of diverse and real landing system failures within a practical budget. GARL employs GA for exploring various environment setups offline, reducing the complexity of RL's online testing in simulating challenging landing scenarios. Our approach outperforms existing methods by up to 18.35% in violation rate and 58% in diversity metric. We validate most discovered violation types with real-world UAV tests, pioneering the integration of offline and online testing strategies for autonomous systems. This method opens new research directions for online testing, with our code and supplementary material available at https://github.com/lfeng0722/drone_testig/.

Efficient sensing is critical for Internet of Things (IoT) applications, such as environmental monitoring and traffic management, where high quality sensing data is essential for decision-making. Traditional sensing methods, however, are often plagued by high deployment costs and incomplete data coverage, significantly limiting their practicality. Despite recent progress, these methods continue to face challenges in maintaining data accuracy, ultimately degrading the Quality of Service (QoS) for IoT applications. To address these limitations, we propose ASSTG, a novel framework that combines an Active Sampling strategy with Spatio-Temporal Graph learning to enable efficient and accurate IoT sensing. At its core, AS-STG is designed to minimize the sampling cost while ensuring the accuracy of the data. The framework begins by analyzing historical data to determine the minimum sampling requirements for accurate inference in subsequent time slots. It then constructs a spatio-temporal graph to model the complex relationships between sensing grids, capturing both spatial and temporal dynamics. To supplement the spatio-temporal information and further optimize representations, we introduce two contrastive learning tasks. Leveraging the refined representation, AS-STG strategically selects informationrich regions for sampling, ensuring that even a sparse subset of samples can provide comprehensive coverage of the entire sensing area. Finally, AS-STG employs matrix completion techniques to reconstruct the complete sensing data from these sparse samples. Extensive experiments on real-world datasets demonstrate that AS-STG significantly outperforms baselines in terms of inference accuracy, cost-efficiency, and scalability. By effectively reducing sampling costs without compromising QoS, AS-STG offers a robust and scalable solution for dynamic IoT sensing systems.

Coverage-guided greybox fuzzing (CGF) has emerged as a powerful technique for software vulnerability detection, yet traditional techniques often struggle with the increasing complexity of modern software systems and the vastness of input spaces. Deep neural networks (DNNs) have begun to fundamentally transform CGF by addressing these limitations through automated feature extraction, adaptive input generation, and intelligent path prioritization. However, despite these advancements, critical gaps persist in understanding the state-of-the-art landscape. Existing studies often lack rigorous benchmarks to evaluate scalability and generalizability, fail to address the interpretability of neural-guided decisions, and overlook the integration of emerging paradigms such as large language models (LLMs) and neurosymbolic reasoning. This survey systematically bridges these gaps by providing a comprehensive taxonomy of DNN-driven CGF techniques, analyzing their strengths and limitations across key fuzzing stages—seed generation, selection, and mutation. We find that although DNNs have significantly improved fuzzing efficiency, challenges such as semantically invalid seeds, high computational overhead, and limited cross-domain adaptability remain unresolved. Most importantly, we identify two transformative directions with the potential to redefine CGF: (1) LLM-powered fuzzing , which combines generative AI with domain-specific fine-tuning to produce context-aware inputs; and (2) neurosymbolic integration , which merges the precision of symbolic execution with the scalability of neural networks to tackle path explosion. By synthesizing these insights, this survey not only clarifies the state-of-the-art but also outlines a roadmap for developing robust, explainable, and widely applicable intelligent fuzzers. The future of CGF lies in hybrid models that integrate data-driven learning with formal methods, paving the way for autonomous vulnerability discovery in an era of increasingly complex software systems.

In the era where Web3.0 values data security and privacy, adopting groundbreaking methods to enhance privacy in recommender systems is crucial. Recommender systems need to balance privacy and accuracy, while also having the ability to overcome cold start problems. The Differential Trust Mechanism (DTM) introduced in this paper is such an approach. The DTM provides a unique use of Gaussian distributions in modeling trust relationships within data, offering a novel way to balance recommendation accuracy with user privacy. This mechanism innovatively applies differential privacy principles, using Gaussian noise addition to protect individual user data from inference attacks, while maintaining the integrity and utility of the overall dataset. Unlike traditional anonymization techniques that often compromise data utility or vulnerability to reverse engineering, DTM provides a robust solution by dynamically adjusting privacy levels based on the trustworthiness of data requests. By combining DTM with existing mainstream recommendation algorithms, the prediction accuracy of MAE and RMSE increases by at least 6.60% and 2.69%, respectively. This dual benefit positions DTM as a significant advancement in secure data processing, especially relevant for online businesses and platforms where personalized recommendations are crucial yet privacy concerns are paramount.

In smart cities, Multi-Camera Multi-Target pedestrian tracking and Re-identification (MCMT-ReID) is essential for effective surveillance, particularly in real-time scenarios, as it demands significant computational resources. Current edgecloud collaboration methods encounter issues such as high latency and potential data leakage due to the physical distance between cloud servers and cameras. To address these issues, we propose a novel Edge-to-Edge Collaboration (E2EC) system that fully utilizes collaboration between heterogeneous edge devices. E2EC partitions the MCMT-ReID task into two modular applications: Tracking and Re-identification (ReID), and employs a customized Kafka communication protocol to optimize data exchange efficiency. Moreover, E2EC dynamically orchestrates intermediate inference flows and transmits features instead of pedestrian detection frames to avoid data leakage. To enhance ReID accuracy, we introduce a real-time ReID Loop Confirmation (ReLC) algorithm, which continuously validates identities to boost reliability and accuracy. E2EC has been deployed and tested in a real-world campus environment to validate its effectiveness. Experimental results demonstrate that E2EC enhances the Rank-1 accuracy and mAP of pedestrian ReID by 36.88% and 46.00%, respectively. Furthermore, it achieves an increase of about 6.35%-12.66% in throughput and reduces latency by 35.01%-57.83% compared to baselines, ensuring realtime performance under dynamic workloads.

The proliferation of contaminated data on Internet of Things (IoT) devices has the potential to undermine the accuracy of data-driven decision-making by altering the distribution of original data. Existing data cleaning methods primarily depend on cloud center or cloud-edge cooperation, leading to prolonged data transmission delays and reduced cleaning accuracy. In this study, we identify edge server placement as a crucial step aligned with data cleaning and view the collaborative edge server placement with distributed data cleaning (SPDC) as a holistic problem. We comprehensively quantify the complexity of our issue through the analysis of numerous scenarios. To address this problem, we introduce a novel distributed collaborative edge framework comprising two key stages: server placement and data cleaning. We propose an optimized clustering algorithm for the former, considering the data distribution on the IoT layer and the constraints of the edge layer. For the latter, we introduce a gossip-based data cleaning algorithm that fully utilizes edge collaboration to enhance data cleaning accuracy. The algorithm exhibits an approximate performance complexity of O($\ln m$), where$m$represents the number of users’ tasks. Both theoretical analysis and experimental results reveal that our algorithm an average improvement in data cleaning accuracy of 9.02% and a reduction in delay of 36.61%, surpassing the performance of state-of-the-art works in various scenarios.

Recent incidents with autonomous vehicles highlight the need for rigorous testing to ensure safety and robustness. Constructing test scenarios for autonomous driving systems (ADSs), however, is labor-intensive. We propose TARGET, an end-to-end framework that automatically generates test scenarios from traffic rules. To address complexity, we leverage a Large Language Model (LLM) to extract knowledge from traffic rules. To mitigate hallucinations caused by large context during input processing, we introduce a domain-specific language (DSL) designed to be syntactically simple and compositional. This design allows the LLM to learn and generate test scenarios in a modular manner while enabling syntactic and semantic validation for each component. Based on these validated representations, TARGET synthesizes executable scripts to render scenarios in simulation. Evaluated seven ADSs with 284 scenarios derived from 54 traffic rules, TARGET uncovered 610 rule violations, collisions, and other issues. For each violation, TARGET generates scenario recordings and detailed logs, aiding root cause analysis. Two identified issues were confirmed by ADS developers: one linked to an existing bug report and the other to limited ADS functionality.

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.

In cloud computing, user requests sent to centralized servers often encounter delay due to network unpredictability, impacting the Quality of Service (QoS) for time-sensitive applications. We propose edge collaboration, utilizing the coordination of edge nodes within regions to handle requests more efficiently and reduce latency. However, edge nodes across different regions struggle with lack of immediate data on resources cached elsewhere, complicating inter-regional request scheduling. To address it, we introduce a federated digital twin model that creates a network linking edge nodes to reflect and update resource statuses in real time. Additionally, we refine the Dijkstra algorithm to optimize routing to the nearest edge nodes based on current network conditions, thereby minimizing delay. Our analyses show that our method significantly lowers delay, enhancing effectiveness over baseline methods.

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.

In beyond 5G (B5G) Internet of Things (IoT) system based on edge intelligence, pay-for-use demand has become a consensus, and the pricing of IoT services has attracted the attention of academia and industry. The pricing method based on noncooperative game allows edge service providers (ESPs) to compete fairly, effectively preventing edge nodes from malicious bidding. However, since only one winner can make a profit each time, it is easy to cause a large number of ESPs to lose money for a long time. To this end, a dynamic alliance pricing mechanism based on distributed edge intelligence is proposed. ESPs can freely choose to form an edge dynamic alliance, which not only retains the independence of edge nodes but also makes full use of the advantages of mutual cooperation between nodes. According to the characteristics of edge nodes, various roles are reasonably divided. In order to prevent abnormal behaviors of edge nodes, we set up necessary restrictive rules. At the same time, we designed a privacy-enhanced joint pricing prediction algorithm to screen the alliance’s candidate solutions to improve pricing efficiency and edge benefit. The experimental results show that, compared with the traditional alliance game method, the performance of the mechanism we proposed improves the utilization rate of edge resources by 32.76%–61.37%. Meanwhile, the prediction accuracy was improved by 16.47%–38.86%, and the average prediction time was reduced by 42.81%–65.57%.

Digital watermarking serves as an effective approach for safeguarding speech signal copyrights, achieved by the incorporation of ownership information into the original signal and its subsequent extraction from the watermarked signal. While traditional watermarking methods can embed and extract watermarks successfully when the watermarked signals are not exposed to severe alterations, these methods cannot withstand attacks such as de-synchronization. In this work, we introduce a novel transformer-based framework designed to enhance the imperceptibility and robustness of speech watermarking. This framework incorporates encoders and decoders built on multi-scale transformer blocks to effectively capture local and long-range features from inputs, such as acoustic features extracted by Short-Time Fourier Transformation (STFT). Further, a deep neural networks (DNNs) based generator, notably the Transformer architecture, is employed to adaptively embed imperceptible watermarks. These perturbations serve as a step for simulating noise, thereby bolstering the watermark robustness during the training phase. Experimental results show the superiority of our proposed framework in terms of watermark imperceptibility and robustness against various watermark attacks. When compared to the currently available related techniques, the framework exhibits an eightfold increase in embedding rate. Further, it also presents superior practicality with scalability and reduced inference time of DNN models.

With the growing popularity of Android devices, Android malware is seriously threatening the safety of users. Although such threats can be detected by deep learning as a service (DLaaS), deep neural networks as the weakest part of DLaaS are often deceived by the adversarial samples elaborated by attackers. In this paper, we propose a new semi-black-box attack framework called one-feature-each-iteration (OFEI) to craft Android adversarial samples. This framework modifies as few features as possible and requires less classifier information to fool the classifier. We conduct a controlled experiment to evaluate our OFEI framework by comparing it with the benchmark methods JSMF, GenAttack and pointwise attack. The experimental results show that our OFEI has a higher misclassification rate of 98.25%. Furthermore, OFEI can extend the traditional white-box attack methods in the image field, such as fast gradient sign method (FGSM) and DeepFool, to craft adversarial samples for Android. Finally, to enhance the security of DLaaS, we use two uncertainties of the Bayesian neural network to construct the combined uncertainty, which is used to detect adversarial samples and achieves a high detection rate of 99.28%.

Federated learning (FL) is a distributed framework for machine learning that enables collaborative training of a shared model across data silos while preserving data privacy. However, the FL aggregation server faces a challenge in waiting for a large volume of model parameters from selected nodes before generating a global model, which leads to inefficient communication and aggregation. Although transmitting only the signs of stochastic gradient descent (SignSGD) reduces the transmission load, it decreases model accuracy, and the time waiting for local model collection remains substantial. Moreover, the security of FL is severely compromised by prevalent poisoning, backdoor, and DDoS attacks, causing ineffective and inaccurate model training. To overcome these challenges, this paper proposes aBlockchain-basedAsynchronousSignSGD (BASS) architecture for efficient and secure federated learning. By integrating a blockchain-based semi-asynchronous aggregation scheme with sign-based gradient compression, BASS considerably improves communication and aggregation efficiency, while providing resistance against attacks. Besides, a novel node-summarized sign aggregation algorithm is developed for the blockchain leaders to ensure the convergence and accuracy of the global model. An open-source prototype is developed, on top of which extensive experiments are conducted. The results validate the superiority of BASS in terms of efficiency, model accuracy, and security.

With the rapid development of intelligent transportation systems (ITS), more and more intelligent applications for ITS have received widespread attention, such as the vehicle detection, inference of typical routes, and traffic forecasting. In these applications, deep learning is widely used as a key artificial intelligence technology. However, most ITS providers fail to collect enough labeled traffic data for model training. As a complement to deep learning, transfer learning is an effective way to solve the scarcity of labeled data, which can transfer knowledge from labeled datasets to unlabeled datasets, thus improving the accuracy of prediction and classification. Nevertheless, when the labeled dataset and the unlabeled dataset are held by different entities, it is still unrealistic for two mutually distrustful entities to cooperate in transfer learning regarding data security and privacy preservation. Although some existing works provide privacy-preserving transfer learning methods, such methods fail to apply to traffic data with high sample dimensions due to their high computational cost and round complexity. To address this problem, we design an efficient privacy-preserving distributed transfer learning protocol, which is appropriate for traffic data. Compared to existing works, our protocol addresses the privacy-preserving problem of transfer learning for traffic data with high sample dimensions. In addition, our protocol has fewer interaction rounds and can be proved in the semi-honest model. Finally, we validate the effectiveness, efficiency and security of the proposed protocol via experiments. Furthermore, we show the application of the proposed protocol in intelligent transportation systems.

Recommendation systems powered by artificial intelligence (AI) are widely used to improve user experience. However, AI inevitably raises privacy leakage and other security issues due to the utilization of extensive user data. Addressing these challenges can protect users’ personal information, benefit service providers, and foster service ecosystems. Presently, numerous techniques based on differential privacy have been proposed to solve this problem. However, existing solutions encounter issues such as inadequate data utilization and a tenuous trade-off between privacy protection and recommendation effectiveness. To enhance recommendation accuracy and protect users’ private data, we propose ID-SR, a novel privacy-preserving social recommendation scheme for trustworthy AI based on the infinite divisibility of Laplace distribution. We first introduce a novel recommendation method adopted in ID-SR, which is established based on matrix factorization with a newly designed social regularization term for improving recommendation effectiveness. We then propose a differential privacy-preserving scheme tailored to the above method that leverages the Laplace distribution’s characteristics to safeguard user data. Theoretical analysis and experimentation evaluation on two publicly available datasets demonstrate that our scheme achieves a superior balance between privacy protection and recommendation effectiveness, ultimately delivering an enhanced user experience.

Federated learning (FL) enables collaborative training of a shared model on edge devices while maintaining data privacy. FL is effective when dealing with independent and identically distributed (iid) datasets, but struggles with non-iid datasets. Various personalized approaches have been proposed, but such approaches fail to handle underlying shifts in data distribution, such as data distribution skew commonly observed in real-world scenarios (e.g., driver behavior in smart transportation systems changing across time and location). Additionally, trust concerns among unacquainted devices and security concerns with the centralized aggregator pose additional challenges. To address these challenges, this paper presents a dynamically optimized personal deep learning scheme based on blockchain and federated learning. Specifically, the innovative smart contract implemented in the blockchain allows distributed edge devices to reach a consensus on the optimal weights of personalized models. Experimental evaluations using multiple models and real-world datasets demonstrate that the proposed scheme achieves higher accuracy and faster convergence compared to traditional federated and personalized learning approaches.

To detect software vulnerabilities with better performance, deep neural networks (DNNs) have received extensive attention recently. However, these vulnerability detection DNN models trained with code representations are vulnerable to specific perturbations on code representations. This motivates us to rethink the bane of software vulnerability detection and find function-agnostic features during code representation which we name as semantic redundant features. This paper first identifies a tight correlation between function-agnostic triggers and semantic redundant feature space (where the redundant features reside) in these DNN models. For correlation identification, we propose a novel Backdoor-based Semantic Redundancy Exploration (BSemRE) framework. In BSemRE, the sensitivity of the trained models to function-agnostic triggers is observed to verify the existence of semantic redundancy in various code representations. Specifically, acting as the typical manifestations of semantic redundancy, naming conventions, ternary operators and identically-true conditions are exploited to generate function-agnostic triggers. Extensive comparative experiments on 1,613,823 samples of eight representative vulnerability datasets and state-of-the-art code representation techniques and vulnerability detection models demonstrate that the existence of semantic redundancy determines the upper trustworthiness limit of DNN-based software vulnerability detection. To the best of our knowledge, this is the first work exploring the bane of software vulnerability detection using backdoor triggers.

Vulnerability detection is a critical problem in software security and attracts growing attention both from academia and industry. Traditionally, software security is safeguarded by designated rule-based detectors that heavily rely on empirical expertise, requiring tremendous effort from software experts to generate rule repositories for large code corpus. Recent advances in deep learning, especially Graph Neural Networks (GNN), have uncovered the feasibility of automatic detection of a wide range of software vulnerabilities. However, prior learning-based works only break programs down into a sequence of word tokens for extracting contextual features of codes, or apply GNN largely on homogeneous graph representation (e.g., AST) without discerning complex types of underlying program entities (e.g., methods, variables). In this work, we are one of the first to explore heterogeneous graph representation in the form of Code Property Graph and adapt a well-known heterogeneous graph network with a dual-supervisor structure for the corresponding graph learning task. Using the prototype built, we have conducted extensive experiments on both synthetic datasets and real-world projects. Compared with the state-of-the-art baselines, the results demonstrate superior performance in vulnerability detection (average F1 improvements over 10% in real-world projects) and language-agnostic transferability from C/C \({+}{+}\) to other programming languages (average F1 improvements over 11%).

The Boolean functions satisfying secure properties on the restricted sets of inputs are studied recently due to their importance in the framework of the FLIP stream cipher. However, finding Boolean functions with optimal cryptographic properties is an open research problem in the cryptographic community. This paper presents an Improved Genetic Algorithm (IGA) with the directed changes that keep the weightwise balancedness of Boolean functions. A cross-protection strategy is proposed to ensure that the offspring has the same weightwise balancedness characteristics of the parents while implementing crossover. Then, a large number of weightwise (almost) perfectly balanced (W(A)PB) functions with a good nonlinearity profile are obtained based on IGA. Finally, we make comparisons between our constructions and relevant works. The comparisons show that IGA has a significant advantage for reaching the W(A)PB functions with high weightwise nonlinearity. Moreover, it is the first time to obtain the 8-variable WPB functions with the weightwise nonlinearity of 28 in the restricted sets of inputs with Hamming weight of 4, and list the statistical indicators of the weightwise nonlinearity for W(A)PB functions for input size n = 9, 10.

The proliferation of dirty data on Internet of Things (IoT) devices can undermine the accuracy of data-driven decision-making by affecting the distribution of original data. The Quality of Service (QoS) of data cleaning on these devices is heavily impacted by processing delay and accuracy. In this paper, we find that edge service placement is a key step aligned with data cleaning and consider the collaborative edge service placement with distributed data cleaning (SPDC) problem. To address this issue, we propose a novel distributed collaborative edge-based architecture that effectively balances the demands of storage, communication, computation, and load constraints. Experimental results show that the proposed approach significantly improves the accuracy of data cleaning by 0.31%-86.07% and reduces delay by 2.73%-58.71% compared to state-of-the-art baselines.

This paper introduces a novel hybrid simulation system (HybriSim) tailored for simulating distributed learning in mobile settings, such as those involving vehicles and pedestrians navigating through cities. Designed to be learning-method independent, the system is compatible with decentralized learning, federated learning, or a combination of the two. It has special relevance for decentralized learning systems that are sensitive to mobility patterns and rely on direct, device-to-device communication. Existing tools for evaluating resource-intensive tasks in opportunistic networks are either purely simulated, which may not accurately reflect system performance, or take the form of testbeds of real devices, which are difficult to scale to use cases involving huge numbers of devices, such as distributed learning. By integrating real devices with virtual simulated devices, HybriSim more accurately mirrors real-world performance and dynamics. This integration not only mitigates the biases associated with pure simulations but also resolves the deployment complexities of conducting simulations entirely on real devices. Our system sets a new benchmark for academic and industry researchers, facilitating more reliable and actionable insights into distributed learning systems in mobility contexts.

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.

The susceptibility of Deep Neural Networks (DNNs) to adversarial attacks has raised concerns regarding their practical applications in real-world scenarios. Although the vulnerability of DNNs to adversarial attacks has been extensively studied in the image domain, research in the audio domain, particularly in the black-box setting with Automatic Speech Recognition (ASR) models, remains limited. While various black-box attacks have been proposed for ASR models, such as transfer attacks, hardware attacks, and query-based attacks, this study concentrates on query-based black-box attacks. The article introduces a new gradient estimation technique, Temporal Natural Evolution Strategies (T-NES), to generate adversarial audio samples more efficiently than existing attacks. T-NES leverages the temporal correlation present in audio to speed up gradient estimation based on the probability scores returned by the target model. The empirical results on benchmark datasets, LibriSpeech and TEDLIUM, and two state-of-the-art ASR models, DeepSpeech2 and Wav2Letter, demonstrate that T-NES can generate successful attacks with up to 30% fewer queries than existing attacks within 500 queries. T-NES could provide a robust baseline for evaluating the black-box adversarial vulnerability of ASR systems.

Cloud and edge computing are gradually used to achieve complex energy operation control and massive information processing in conventional power grid. Meanwhile, with the tremendous number of distributed energy resources and power equipment integrated into the smart grid enabled by cloud and edge computing, the centralized distribution network cannot realize the flexible and realtime energy supply due to the unpredictable and wide distribution of distributed energy resources, which will further deteriorate the stability of smart grid. To solve those problems, this article proposes energy centric smart grid to achieve power dispatching with the help of cloud-edge computing. Our solution uses energy caching and energy multiple addressing of the edge router to eliminate the intermittency of renewables and speed up energy response. For the stability of energy market and to encourage users to participate in power dispatching, a cloud-edge computing-driven energy cache orchestration mechanism is designed. The empirical results show that the response time is greatly reduced to meet the stringent quality of service requirement in smart grid integrated with distributed energy resources.

With the increasing penetration of the Internet of things (IoT) into people's lives, the limitations of traditional medical systems are emerging. First, the typical way of handling sensitive information can easily lead to privacy disclosure. Second, the medical system is relatively isolated. It is difficult for one medical system to share data with another, and the scope of users' activities is limited within the system boundary. To solve these two problems, we propose a new privacy-preserving medical data-sharing scheme by introducing the authorization mechanism and attribute-based encryption (ABE) based on blockchain, which breaks system boundaries and realizes data sharing among several medical institutions. ABE is used to realize scalable access control. In addition, doctors can share their knowledge to diagnose users by introducing many-to-many matching, which means that patients' health data can be represented by multiple keywords and doctors' expertise can be represented by multiple interests. We provide the correctness and security analysis of our scheme and implement a prototype tool on Ethereum. The experimental results show that our scheme solves the contradiction between the privacy preservation of medical data and the necessity of data sharing.

Recently, wireless edge networks have realized intelligent operation and management with edge artificial intelligence (AI) techniques (i.e., federated edge learning). However, the trustworthiness and effective incentive mechanisms of federated edge learning (FEL) have not been fully studied. Thus, the current FEL framework will still suffer untrustworthy or low-quality learning parameters from malicious or inactive learners, which undermines the viability and stability of FEL. To address these challenges, the potential social attributes among edge devices and their users can be exploited, while not included in previous works. In this paper, we propose a novelSocialFederatedEdgeLearning framework (SFEL) over wireless networks, which recruits trustworthy social friends as learning partners. First, we build a social graph model to find like-minded friends, comprehensively considering the mutual trust and learning task similarity. Besides, we propose a social effect based incentive mechanism for better personal federated learning behaviors with both complete and incomplete information. Finally, we conduct extensive simulations with the Erdos-Renyi random network, the Facebook network, and the classic MNIST/CIFAR-10 datasets. Simulation results demonstrate our framework could realize trustworthy and efficient federated learning over wireless edge networks, and it is superior to the existing FEL incentive mechanisms that ignore social effects.

Machine learning has made significant progress in image recognition, natural language processing, and autonomous driving. However, the generation of adversarial examples has proved that the machine learning system is unreliable. By adding imperceptible perturbations to clean images can fool the well-trained machine learning systems. To solve this problem, we propose an adaptive image denoising framework Adaptive Scalar Quantization (ASQ-FastBM3D). TheASQ-FastBM3Dframework combines theASQmethod with theFastBM3Dalgorithm. The adaptive scalar quantization is the improvement of scalar quantization, which is used to eliminate most of the perturbations.FastBM3Dis proposed to improve the quality of the quantified image. The running time ofFastBM3Dis 50% less than that ofBM3D. Compared with some traditional filter methods and some state-of-the-art neural network methods for recovering the adversarial examples, the accuracy rate of ourASQ-FastBM3Dmethod is 99.73% and the F1 score is 98.01%, which is the highest.

Autonomous driving has gained much attention from both industry and academia. Currently, Deep Neural Networks (DNNs) are widely used for perception and control in autonomous driving. However, several fatal accidents caused by autonomous vehicles have raised serious safety concerns about autonomous driving models. Some recent studies have successfully used the metamorphic testing technique to detect thousands of potential issues in some popularly used autonomous driving models. However, prior study is limited to a small set of metamorphic relations, which do not reflect rich, real-world traffic scenarios and are also not customizable. This paper presents a novel declarative rule-based metamorphic testing framework calledRMT.RMTprovides a rule template with natural language syntax, allowing users to flexibly specify an enriched set of testing scenarios based on real-world traffic rules and domain knowledge.RMTautomatically parses human-written rules to metamorphic relations using an NLP-based rule parser referring to an ontology list and generates test cases with a variety of image transformation engines. We evaluatedRMTon three autonomous driving models. With an enriched set of metamorphic relations,RMTdetected a significant number of abnormal model predictions that were not detected by prior work. Through a large-scale human study on Amazon Mechanical Turk, we further confirmed the authenticity of test cases generated byRMTand the validity of detected abnormal model predictions.

With the rapid development of wireless sensor networks (WSNs) and the Internet of Things (IoT), increasing computing tasks are sinking to mobile edge networks, such as distributed learning systems. These systems benefit from the massive amounts of data and computing power on mobile devices and can learn qualified models on the premise of protecting user privacy. In fact, coordinating mobile devices to participate in computing is challenging. On the one hand, the heterogeneous performance of devices makes it difficult to guarantee computing efficiency. On the other hand, there are unreliable factors in the mobile network, which will destroy the stability of the distributed learning. Therefore, we design a three-layer framework called an edge-intelligence-based distributed learning system (EIDLS). Specifically, a novel multilayer perceptron-based device availability evaluation model is proposed to select devices with good performance. The evaluation model performs online learning and optimization according to the resources (CPU, battery, etc.) of devices. Meanwhile, we propose a dynamic trust evaluation algorithm to reduce the side effects of unreliable devices. The experimental results of some commonly used datasets validate that the proposed EIDLS dramatically minimizes the energy consumption and communication cost and improves the calculation accuracy and the stability of the system.

Sleep stage recognition is crucial for assessing sleep and diagnosing chronic diseases. Deep learning models, such as Convolutional Neural Networks and Recurrent Neural Networks, are trained using grid data as input, making them not capable of learning relationships in non-Euclidean spaces. Graph-based deep models have been developed to address this issue when investigating the external relationship of electrode signals across different brain regions. However, the models cannot solve problems related to the internal relationships between segments of electrode signals within a specific brain region. In this study, we propose a Pearson correlation-based graph attention network, called PearNet, as a solution to this problem. Graph nodes are generated based on the spatial-temporal features extracted by a hierarchical feature extraction method, and then the graph structure is learned adaptively to build node connections. Based on our experiments on the Sleep-EDF-20 and Sleep-EDF-78 datasets, PearNet performs better than the state-of-the-art baselines.

When developing autonomous driving systems (ADS), developers often need to replay previously collected driving recordings to check the correctness of newly introduced changes to the system. However, simply replaying the entire recording is not necessary given the high redundancy of driving scenes in a recording (e.g., keeping the same lane for 10 minutes on a highway). In this pa- per, we propose a novel test reduction and prioritization approach for multi-module ADS. First, our approach automatically encodes frames in a driving recording to feature vectors based on a driving scene schema. Then, the given recording is sliced into segments based on the similarity of consecutive vectors. Lengthy segments are truncated to reduce the length of a recording and redundant segments with the same vector are removed. The remaining seg- ments are prioritized based on both the coverage and the rarity of driving scenes. We implemented this approach on an industry- level, multi-module ADS called Apollo and evaluated it on three road maps in various regression settings. The results show that our approach significantly reduced the original recordings by over 34% while keeping comparable test effectiveness, identifying almost all injected faults. Furthermore, our test prioritization method achieves about 22% to 39% and 41% to 53% improvements over three baselines in terms of both the average percentage of faults detected (APFD) and TOP-K.

Autonomous driving has shown great potential to reform modern transportation. Yet its reliability and safety have drawn a lot of attention and concerns. Compared with traditional software systems, autonomous driving systems (ADSs) often use deep neural networks in tandem with logic-based modules. This new paradigm poses unique challenges for software testing. Despite the recent development of new ADS testing techniques, it is not clear to what extent those techniques have addressed the needs of ADS practitioners. To fill this gap, we present the first comprehensive study to identify the current practices and needs of ADS testing. We conducted semi-structured interviews with developers from 10 autonomous driving companies and surveyed 100 developers who have worked on autonomous driving systems. A systematic analysis of the interview and survey data revealed 7 common practices and 4 emerging needs of autonomous driving testing. Through a comprehensive literature review, we developed a taxonomy of existing ADS testing techniques and analyzed the gap between ADS research and practitioners’ needs. Finally, we proposed several future directions for SE researchers, such as developing test reduction techniques to accelerate simulation-based ADS testing.

In the era of the sixth generation (6G), the deployment of massive Internet of Things (IoT) generates and processes large amounts of data, resulting in high energy demand and huge challenges to the energy-limited IoT devices. To achieve green and sustainable communication, energy harvesting is a feasible technology to prolong the lifetime of IoT. However, the existing energy harvesting architecture cannot guarantee the privacy of energy users while improving the intelligence and effectiveness of energy transmission. To solve these issues, we propose a differential privacy and intelligent reflecting surface empowered privacy-preserving energy harvesting framework for 6G-enabled IoT. First, a secure and intelligent energy harvesting framework is designed, which includes an intelligent reflecting surface-aided radio frequency power transmission mechanism and a differential privacy-based energy harvesting mechanism. Second, an exponential mechanism-based privacy-preserving energy harvesting scheme is established, where we analyze the adversary mode, propose the differential privacy-enabled location-preserving algorithm, and provide security analysis and proof. Third, we quantify the user satisfaction for energy harvesting and propose a deep reinforcement learning empowered resource allocation scheme to maximize the weighted satisfaction of all system users. Finally, simulation results show the effectiveness of the proposed secure and intelligent energy harvesting architecture for 6G IoT.

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.

Data dissemination of vehicles is critical for vehicular networks because of the extensive impact of traffic information. The existing works for data dissemination in vehicular networks mainly use data scheduling algorithms to transmit data among vehicles. However, it is challenging to meet the ultrareliable and low-latency requirements of data transmission among vehicular networks due to the intrinsic movement characteristic of vehicles. To promote the data dissemination of vehicular networks, a data dissemination algorithm with trajectory privacy protection is proposed in this article, which leverages the cooperative distribution of key tasks and distance deviation. Specifically, the key tasks are disseminated to vehicles first, and the trajectory privacy protection scheme is further developed to guarantee the security of data transmission by the exploration of distance deviation and pseudonym entropy. Simulation results indicate that the proposed adaptive data dissemination algorithm is approximately 60%, 69%, and 50% better than the state-of-the-art scheduling algorithms in terms of connectivity degree, transmission delay, and average distance deviation for the vehicular network.

The lightweight privacy-preserving algorithm in the vehicular networks (VNs) improves the reliability of data transmission for the vehicles. However, it is challenging for vehicles to execute resource-consuming algorithms while driving. In addition, the high-speed mobility of vehicles also brings data scheduling problems for vehicles and other equipment. To tackle the problems mentioned above, this article proposes a privacy-preserving data scheduling in an incentive-driven VN, which achieves efficient and secure data transmission based on an incentive mechanism among vehicles. The algorithm first balances the benefits between the source, forwarding, and destination nodes through a multidimensional incentive mechanism to ensure positive benefits. After the vehicles participate in the task under the action of the incentive mechanism, the key task will then be identified and completed. Finally, the data interference mechanism guarantees the security of data transmission between the vehicle and the edge server. The simulation experiment results show that the proposed algorithm is superior to other algorithms in revenue, satisfaction, and reliability.

Federated learning (FL) is an emerging paradigm through which decentralized devices can collaboratively train a common model. However, a serious concern is the leakage of privacy from exchanged gradient information between clients and the parameter server (PS) in FL. To protect gradient information, clients can adopt differential privacy (DP) to add additional noises and distort original gradients before they are uploaded to the PS. Nevertheless, the model accuracy will be significantly impaired by DP noises, making DP impracticable in real systems. In this work, we propose a novel noise information secretly sharing (NISS) algorithm to alleviate the disturbance of DP noises by sharing negated noises among clients. We theoretically prove that: 1) if clients are trustworthy, DP noises can be perfectly offset on the PS and 2) clients can easily distort negated DP noises to protect themselves in case that other clients are not totally trustworthy, though the cost lowers model accuracy. NISS is particularly applicable for FL across multiple Internet of Things (IoT) systems, in which all IoT devices need to collaboratively train a model. To verify the effectiveness and the superiority of the NISS algorithm, we conduct experiments with the MNIST and CIFAR-10 data sets. The experimental results verify our analysis and demonstrate that NISS can improve model accuracy by 19% on average and obtain better privacy protection if clients are trustworthy.

To reduce uploading bandwidth and address privacy concerns, deep learning at the network edge has been an emerging topic. Typically, edge devices collaboratively train a shared model using real-time generated data through the Parameter Server framework. Although all the edge devices can share the computing workloads, the distributed training processes over edge networks are still time-consuming due to the parameters and gradients transmission procedures between parameter servers and edge devices. Focusing on accelerating distributed Convolutional Neural Networks (CNNs) training at the network edge, we present DynaComm, a novel scheduler that dynamically decomposes each transmission procedure into several segments to achieve optimal layer-wise communications and computations overlapping during run-time. Through experiments, we verify that DynaComm manages to achieve optimal layer-wise scheduling for all cases compared to competing strategies while the model accuracy remains untouched.

While promoting the intelligence in industrial production, Industry 4.0 has also caused privacy leaks concurrently. As a possible solution, the existing personalized federated learning relies too much on a good global model to fine-tune or limit local drift, which lacks intelligent cognitive ability. When faced with heterogeneous data or poisoning attacks, even a few low-quality local models will affect the whole federation effect. In this article, we design a cognitive personalized federated learning (CPFL) mechanism for Industry 4.0, which can selectively improve the collaboration capabilities between more relevant devices. We use the parameters in the local training process as the cognitive basis and calculate Earth mover’s distance to quantify the differences between different models. When the gradient distribution is closer, the local data are more similar. By adaptively adjusting the weight distribution during the aggregation process, self-learning and cooperative learning are balanced, and the interference of heterogeneous data on the federated training process is reduced. Therefore, the global model can better fit most heterogeneous industrial data and achieve personalization. Comparative experimental results show that our proposed CPFL mechanism can increase the average accuracy of personalized models by 5%–10% in non independent and identically distributed situations, and it has certain effects against poisoning attacks and noise interference.

Negative mood states include tension, depression, anger, fatigue, and confusion, which represent the weak internal emotions of a human. Negative mood states exert adverse impact on individuals' ability to make rational decisions, which entails the practicable method of negative mood state detection. The most commonly used negative mood state detection methods are based on the psychological scale, which requires additional work and brings inconvenience to the subject in the application scenarios. To overcome this challenge, this paper proposes a novel non-contact negative mood state detection method according to the knowledge of affective computing. The POMS-net model is used to extract temporal-spatial features from visible and infrared thermal videos, and the negative mood state detection is realized using data reliability-focused multi-modal fusion. The proposed method is verified using the HDT-BR dataset collected in the aerospace medicine experiment "Earth-Star II" and the VIRI public dataset. The experimental results on the datasets verify that our method outperforms the comparison methods.

With recent developments in communication technologies, vehicular networks have become a reality with various applications. However, the cybersecurity aspect of vehicular networks is still an open issue that needs to be addressed with novel defence mechanisms against attacks. This paper first presents the state-of-the-art communication technologies in vehicular networks (either inter-vehicle networking or in-vehicle networking) along with their applications. Then we explore novel technologies including machine learning and blockchain as cybersecurity defence mechanisms in vehicular networks. Based on the extensive survey, we highlight some insights for future research to secure vehicular networks.

It is always a challenging problem to deliver a huge volume of videos over the Internet. To meet the high bandwidth and stringent playback demand, one feasible solution is to cache video contents on edge servers based on predicted video popularity. Traditional caching algorithms (e.g., LRU, LFU) are too simple to capture the dynamics of video popularity, especially long-tailed videos. Recent learning-driven caching algorithms (e.g., DeepCache) show promising performance, however, such black-box approaches are lack of explainability and interpretability. Moreover, the parameter tuning requires a large number of historical records, which are difficult to obtain for videos with low popularity. In this paper, we optimize video caching at the edge using a white-box approach, which is highly efficient and also completely explainable. To accurately capture the evolution of video popularity, we develop a mathematical model calledHRSmodel, which is the combination of multiple point processes, including Hawkes’ self-exciting, reactive and self-correcting processes. The key advantage of the HRS model is its explainability, and much less number of model parameters. In addition, all its model parameters can be learned automatically through maximizing the Log-likelihood function constructed by past video request events. Next, we further design an online HRS-based video caching algorithm. To verify its effectiveness, we conduct a series of experiments using real video traces collected from Tencent Video, one of the largest online video providers in China. Experiment results demonstrate that our proposed algorithm outperforms the state-of-the-art algorithms, with 15.5% improvement on average in terms of cache hit rate under realistic settings.

Recent years have seen a rapid development of machine learning based multi-module unmanned aerial vehicle (UAV) systems. To address the oracle problem in autonomous systems, numerous studies have been conducted to use metamorphic testing to automatically generate test scenes for various modules, e.g., those in self-driving cars. However, as most of the studies are based on unit testing including end-to-end model-based testing, a similar testing approach may not be equally effective for UAV systems where multiple modules are working closely together. Therefore, in this paper, instead of unit testing, we propose a novel metamorphic system testing framework for UAV, named MSTU, to detect the defects in multi-module UAV systems. A preliminary evaluation plan to apply MSTU on an emerging autonomous multi-module UAV system is also presented to demonstrate the feasibility of the proposed testing framework.

Pervasive computing applications commonly involve user's personal smartphones collecting data to influence application behavior. Applications are often backed by models that learn from the user's experiences to provide personalized and responsive behavior. While models are often pre-trained on massive datasets, federated learning has gained attention for its ability to train globally shared models on users' private data without requiring the users to share their data directly. However, federated learning requires devices to collaborate via a central server, under the assumption that all users desire to learn the same model. We define a new approach, opportunistic federated learning, in which individual devices belonging to different users seek to learn robust models that are personalized to their user's own experiences. However, instead of learning in isolation, these models opportunistically incorporate the learned experiences of other devices they encounter opportunistically. In this paper, we explore the feasibility and limits of such an approach, culminating in a framework that supports encounter-based pairwise collaborative learning. The use of our opportunistic encounter-based learning amplifies the performance of personalized learning while resisting overfitting to encountered data.

With recent advances in autonomous driving, voice control systems have become increasingly adopted as human-vehicle interaction methods. This technology enables drivers to use voice commands to control the vehicle and will be soon available in Advanced Driver Assistance Systems (ADAS). Prior work has shown that Siri, Alexa and Cortana, are highly vulnerable to inaudible command attacks. This could be extended to ADAS in real-world applications and such an inaudible command threat is difficult to detect due to microphone nonlinearities. In this paper, we aim to develop a more practical solution by using camera views to defend against inaudible command attacks where ADAS are capable of detecting their environment via multi-sensors. To this end, we propose a novel multimodal deep learning classification system to defend against inaudible command attacks. Our experimental results confirm the feasibility of the proposed defense methods and the best classification accuracy reaches 89.2%. Code is available at https://github.com/ITSEG-MQ/Sensor-Fusion-Against-VoiceCommand-Attacks.

In federated learning, machine learning and deep learning models are trained globally on distributed devices. The state-of-the-art privacy-preserving technique in the context of federated learning is user-level differential privacy. However, such a mechanism is vulnerable to some specific model poisoning attacks such as Sybil attacks. A malicious adversary could create multiple fake clients or collude compromised devices in Sybil attacks to mount direct model updates manipulation. Recent works on novel defense against model poisoning attacks are difficult to detect Sybil attacks when differential privacy is utilized, as it masks clients' model updates with perturbation. In this work, we implement the first Sybil attacks on differential privacy based federated learning architectures and show their impacts on model convergence. We randomly compromise some clients by manipulating different noise levels reflected by the local privacy budget ε of differential privacy with Laplace mechanism on the local model updates of these Sybil clients. As a result, the global model convergence rates decrease or even leads to divergence. We apply our attacks to two recent aggregation defense mechanisms, called Krum and Trimmed Mean. Our evaluation results on the MNIST and CIFAR-10 datasets show that our attacks effectively slow down the convergence of the global models. We then propose a method to keep monitoring the average loss of all participants in each round for convergence anomaly detection and defend our Sybil attacks based on the training loss reported from randomly selected sets of clients as the judging panels. Our empirical study demonstrates that our defense effectively mitigates the impact of our Sybil attacks.

Nowadays, Deep Learning-as-a-Service can be de-ployed in the Internet of Things (IoT) to provide smart services and sensor data processing. However, recent research has re-vealed that some Deep Neural Networks (DNN) can be easily misled by adding relatively small but adversarial perturbations to the input (e.g., pixel mutation in input images). One challenge in defending DNN against these attacks is to efficiently identify and filtering out the adversarial pixels. The state-of-the-art defense strategies with good robustness often require additional model training for specific attacks. To reduce the computational cost without loss of generality, we present a defense strategy called a progressive defense against adversarial attacks (PDAAA) for efficiently and effectively filtering out the adversarial pixel mutations, which could mislead the neural network towards erro-neous outputs, without a-priori knowledge about the attack type. We evaluated our progressive defense strategy against various attack methods on two well-known datasets. Experimental result shows it outperforms the state-of-the-art methods(Adversarial-PGD, Adversarial-Network, and Adversarial-Dual-Network) with dramatically reduced computation cost.

Federated learning (FL) is a promising new technology in the field of IoT intelligence. However, exchanging model-related data in FL may leak the sensitive information of participants. To address this problem, we propose a novel privacy-preserving FL framework based on an innovative chained secure multiparty computing technique, named chain-PPFL. Our scheme mainly leverages two mechanisms: 1) single-masking mechanism that protects information exchanged between participants and 2) chained-communication mechanism that enables masked information to be transferred between participants with a serial chain frame. We conduct extensive simulation-based experiments using two public data sets (MNIST and CIFAR-100) by comparing both training accuracy and leak defence with other state-of-the-art schemes. We set two data sample distributions (IID and NonIID) and three training models (CNN, MLP, and L-BFGS) in our experiments. The experimental results demonstrate that the chain-PPFL scheme can achieve practical privacy preservation (equivalent to differential privacy with ∈ approaching zero) for FL with some cost of communication and without impairing the accuracy and convergence speed of the training model.

The convergence of AI and IoT enables data to be quickly explored and turned into vital decisions, and however, there are still some challenging issues to be further addressed. For example, lacking of enough data in AI-based decision making [so-called sparse decision making (SDM)] will decrease the efficiency dramatically, or even disable the intelligent IoT networks. Taking the intelligent IoT networks as the network infrastructure, the recommendation systems have been facing such SDM problems. A naive solution is to introduce trust information. However, trust information may also face the difficulty of sparse trust evidence (also known as sparse trust problem). In our work, an accurate SDM model with two-way trust recommendation in the AI-enabled IoT systems is proposed, named TT-SVD. Our model incorporates both trust information and rating information more thoroughly, which can efficiently alleviate the above-mentioned sparse trust problem and therefore be able to solve the cold start and data sparsity problems. Specifically, we first consider the twofold trust influences from both trustees and trusters, which can be represented by a factor named trust propensity. To this end, we propose a dual model, including a truster model (TrusterSVD) and a trustee model (TrusteeSVD) based on an existing rating-only recommendation model called SVD++, which are integrated by the weighted average and yield the final model, TT-SVD. The experimental results show that our model outperforms the state-of-the-art, including SVD and TrustSVD in both the “all users” and “cold start users” cases, and the accuracy improvement can reach a maximum of 29%. Complexity analysis shows that our model is equally suitable for the case of large sparse data sets. In summary, our model can effectively solve the sparse decision problem by introducing the two-way trust recommendation, and hence improve the efficiency of the intelligent recommendation systems.

The emergence of fog computing has brought unprecedented opportunities to the Internet-of-Things (IoT) field, and it is now feasible to incorporate deep learning at the edge of the IoT network to provide a wide range of highly tailored services. In this article, we present a fog-based democratically collaborative learning scheme in which fog nodes collaborate on the model training process even without the support of the cloud, contributing to the advances of IoT in terms of realizing a more intelligent edge. To achieve that, we design a voting strategy so that a fog node could be elected as the coordinator node based on both distance and computational power metrics to coordinate the training process. Also, a collaborative learning algorithm is proposed to generalize the training of different deep learning models in the fog-enabled IoT environment. We then implement two popular use cases, including a user trajectory prediction and a distributed image recognition, to demonstrate the feasibility, practicality, and effectiveness of the scheme. More importantly, the experiments on both use cases are conducted through a real world, in-door fog deployment. The result shows that the scheme can utilize fog to obtain a well-performing deep learning model in the cloudless IoT environment while mitigating the data locality issue for each fog node.

With the extensive application of artificial intelligence technology in 5G and Beyond Fifth Generation (B5G) networks, it has become a common trend for artificial intelligence to integrate into modern communication networks. Deep learning is a subset of machine learning and has recently led to significant improvements in many fields. In particular, many 5G-based services use deep learning technology to provide better services. Although deep learning is powerful, it is still vulnerable when faced with 5G-based deep learning services. Because of the nonlinearity of deep learning algorithms, slight perturbation input by the attacker will result in big changes in the output. Although many researchers have proposed methods against adversarial attacks, these methods are not always effective against powerful attacks such as CW. In this paper, we propose a new two-stream network which includes RGB stream and spatial rich model (SRM) noise stream to discover the difference between adversarial examples and clean examples. The RGB stream uses raw data to capture subtle differences in adversarial samples. The SRM noise stream uses the SRM filters to get noise features. We regard the noise features as additional evidence for adversarial detection. Then, we adopt bilinear pooling to fuse the RGB features and the SRM features. Finally, the final features are input into the decision network to decide whether the image is adversarial or not. Experimental results show that our proposed method can accurately detect adversarial examples. Even with powerful attacks, we can still achieve a detection rate of 91.3%. Moreover, our method has good transferability to generalize to other adversaries.

Nowadays, IoT systems can better satisfy the service requirements of users with effectively utilizing edge computing resources. Designing an appropriate pricing scheme is critical for users to obtain the optimal computing resources at a reasonable price and for service providers to maximize profits. This problem is complicated with incomplete information. The state-of-the-art solutions focus on the pricing game between a single service provider and users, which ignoring the competition among multiple edge service providers. To address this challenge, we design an edge-intelligent hierarchical dynamic pricing mechanism based on cloud-edge-client collaboration. We introduce an improved double-layer Stackelberg game model to describe the cloud-edge-client collaboration. Technically, we propose a novel pricing prediction algorithm based on double-label Radius K-nearest Neighbors, thereby reducing the number of invalid games to accelerate the game convergence. The experimental results show that our proposed mechanism effectively improves the quality of service for users and realizes the maximum benefit equilibrium for service providers, compared with the traditional pricing scheme. Our proposed mechanism is highly suitable for the IoT applications (e.g., intelligent agriculture or Internet of Vehicles), where there are multiple competing edge service providers for resource allocation.

Most existing robust principal component analysis (PCA) and 2-D PCA (2DPCA) methods involving the l2-norm can mitigate the sensitivity to outliers in the domains of image analysis and pattern recognition. However, existing approaches neither preserve the structural information of data in the optimization objective nor have the robustness of generalized performance. To address the above problems, we propose two novel center-weight-based models, namely, centered PCA (C-PCA) and generalized centered 2DPCA with l2,p-norm minimization (GC-2DPCA), which are developed for vector- and matrix-based data, respectively. The C-PCA can preserve the structural information of data by measuring the similarity between the data points and can also retain the PCA's original desirable properties such as the rotational invariance. Furthermore, GC-2DPCA can learn efficient and robust projection matrices to suppress outliers by utilizing the variations between each row of the image matrix and employing power p of l2,1-norm. We also propose an efficient algorithm to solve the C-PCA model and an iterative optimization algorithm to solve the GC-2DPCA model, and we theoretically analyze their convergence properties. Experiments on three public databases show that our models yield significant improvements over the state-of-the-art PCA and 2DPCA approaches.

As recommendation systems continue to evolve, researchers are using trust data to improve the accuracy of recommendation prediction and help users find relevant information. However, large recommendation systems with trust data suffer from the sparse trust problem, which leads to grade inflation and severely affects the reliability of trust propagation. This paper presents a novel research on sparse trust data mining, which includes the new concept of sparse trust, a sparse trust model, and a trust mining framework. It lays a foundation for the trust-related research in large recommended systems. The new trust mining framework is based on customized normalization functions and a novel transitive gossip trust model, which discovers potential trust information between entities in a large-scale user network and applies it to a recommendation system. We conducts a comprehensive performance evaluation on both real-world and synthetic datasets. The results confirm that our framework mines new trust and effectively ameliorates sparse trust problem.

The industrial control and automation systems have played an increasingly important role in critical manufacturing processes. In such systems, many Internet of Things devices continuously collect large number of streaming data for real-time processing. Verifiable data streaming (VDS) addresses such authenticity issue for streaming data, but most VDS schemes are not efficient and lightweight, do not support range querying, and cannot be used in practice. To improve the efficiency and achieve a verifiable range query in data streaming, we present here a new primitive, namely, a chameleon authentication tree with prefixes (PCAT), which is extended from the PBTree and chameleon authentication tree. Our scheme is not only lightweight but also supports dynamic expansion and verifiable range query in data streaming, making it more suitable for resource-constrained devices. We separate the PCAT's algorithms into the following phases: initialization, data appending, query, and verification. Our analyses prove that the PCAT satisfies all the security requirements of VDS. Moreover, an efficiency analysis and performance evaluation demonstrate that our scheme not only supports lightweight data streaming authentication but also has high efficiency, which means that the PCAT is easier to apply in the industrial control and automation systems.

The rapid development of artificial intelligence, especially deep learning technology, has advanced autonomous driving systems (ADSs) by providing precise control decisions to counterpart almost any driving event, spanning from antifatigue safe driving to intelligent route planning. However, ADSs are still plagued by increasing threats from different attacks, which could be categorized into physical attacks, cyberattacks and learning-based adversarial attacks. Inevitably, the safety and security of deep learning-based autonomous driving are severely challenged by these attacks, from which the countermeasures should be analyzed and studied comprehensively to mitigate all potential risks. This survey provides a thorough analysis of different attacks that may jeopardize ADSs, as well as the corresponding state-of-the-art defense mechanisms. The analysis is unrolled by taking an in-depth overview of each step in the ADS workflow, covering adversarial attacks for various deep learning models and attacks in both physical and cyber context. Furthermore, some promising research directions are suggested in order to improve deep learning-based autonomous driving safety, including model robustness training, model testing and verification, and anomaly detection based on cloud/edge servers.

Three-dimensional display and virtual reality technology have been applied in minimally invasive surgery to provide doctors with a more immersive surgical experience. One of the most popular systems based on this technology is the Da Vinci surgical robot system. The key to build the in vivo 3-D virtual reality model with a monocular endoscope is an accurate estimation of depth and motion. In this article, a fully unsupervised learning method for depth and motion estimation using the continuous monocular endoscopic video is proposed. After the detection of highlighted regions, EndoMotionNet and EndoDepthNet are designed to estimate ego-motion and depth, respectively. The timing information between consecutive frames is considered with a long short-term memory layer by EndoMotionNet to enhance the accuracy of ego-motion estimation. The estimated depth value of the previous frame is used to estimate the depth of the next frame by EndoDepthNet with a multimode fusion mechanism. The custom loss function is defined to improve the robustness and accuracy of the proposed unsupervised-learning-based method. Experiments with the public datasets verify that the proposed unsupervised-learning-based continuous depth and motion estimation method can effectively improve the accuracy of depth and motion estimation, especially after processing the frame.

As major infrastructures are increasingly depending on electricity, the smart grid has become an important base for industrial manufacturing and residential living. Despite the benefits of smart grids, the reliability and continuity of power services are often threatened by severe nature disasters and human errors. In smart grids, the centralized and often large-sized grid equipment hinder the rapid recovery and flexible reconfiguration in an emergency. Meanwhile, the large amount of personal equipment and their invisibility make it difficult for the grid operators to utilize assets optimally and easily. In addition, since the power service is provided by multiple energy functions, which consists voltage transformation, transmission, and storage, only considering the restoration of power generation function will restrict the service capacity and lengthen the response time. To address these problems, this article proposes an energy function virtualization for smart grid to decouple the implementation of energy functions from the underlying physical infrastructure to speed up the deployment and test of energy functions. With the help of distributed infrastructure resources, manager can redeploy energy functions and accelerate the service response in smart grid. To motivate prosumers to contribute private function resources, an optimized network calculus performance assessment scheme and a game theory-based resource orchestration scheme are proposed. Simulation results show that proposed scheme can dynamically adjust the delay factor to shorten the emergency response time.

Time-series medical images are an important type of medical data that contain rich temporal and spatial information. As a state-of-the-art, computer-aided diagnosis (CAD) algorithms are usually used on these image sequences to improve analysis accuracy. However, such CAD algorithms are often required to upload medical images to honest-but-curious servers, which introduces severe privacy concerns. To preserve privacy, the existing CAD algorithms support analysis on each encrypted image but not on the whole encrypted image sequences, which leads to the loss of important temporal information among frames. To meet this challenge, a convolutional-LSTM network, named HE-CLSTM, is proposed for analyzing time-series medical images encrypted by a fully homomorphic encryption mechanism. Specifically, several convolutional blocks are constructed to extract discriminative spatial features, and LSTM-based sequence analysis layers (HE-LSTM) are leveraged to encode temporal information from the encrypted image sequences. Moreover, a weighted unit and a sequence voting layer are designed to incorporate both spatial and temporal features with different weights to improve performance while reducing the missed diagnosis rate. The experimental results on two challenging benchmarks (a Cervigram dataset and the BreaKHis public dataset) provide strong evidence that our framework can encode visual representations and sequential dynamics from encrypted medical image sequences; our method achieved AUCs above 0.94 both on the Cervigram and BreaKHis datasets, constituting a significant margin of statistical improvement compared with several competing methods.

With the rapid development of various computing technologies, the constraints of data processing capabilities gradually disappeared, and more data can be simultaneously processed to obtain better performance compared to conventional methods. As a standard statistical analysis method that has been widely used in many fields, Independent Component Analysis (ICA) provides a new way for motion detection by extracting the foreground without precisely modeling the background. However, most existing ICA-based motion detection algorithms use only two-channel data for source separation and simply generate the observation vectors by decomposing and reconstructing the images by row, hence they cannot obtain an integrated and accurate shape of the moving objects in complex scenes. In this article, we propose a refined ICA algorithm for motion detection (RICA-MD), which fuses a larger number of channels than conventional ICA-based motion detection algorithms to provide more effective information for foreground extraction. Meanwhile, we propose four novel methods for generating observation vectors to further cover the diverse motion styles of the moving objects. These improvements enable RICA-MD to effectively deal with slowly moving objects, which are difficult to detect using conventional methods. Our quantitative evaluation in multiple scenes shows that our proposed method is able to achieve a better performance at an acceptable cost of false alarms.

As an ideal infrastructure for smart services, Fog Computing is becoming the next wave of IT investment harnessing the successful models of Cloud Computing and latest technologies such as 5G and Internet of Things (IoT). However, the development of Fog Computing systems is a big challenge due to its complex, heterogeneous and distributed nature. Currently, there are a few SDKs released by some public Cloud service providers to support the development of Fog services in a top-down fashion as the key motive is to leverage their business Cloud services. However, Fog Computing systems are usually designed in a bottom-up fashion as the major functionalities are centred around the Edge Nodes and the End Devices. Meanwhile, significant efforts are required to verify the conformance of software behaviours as the collaboration between the End Devices, Edge Nodes and Cloud Servers is vital to the success of a Fog Computing System. Therefore, a holistically designed software development platform is urgently required. In this paper, we propose TDD4Fog, a test-driven software development platform for Fog Computing systems. Following the Test-Driven Development (TDD) methodology and a bottom-up design fashion, TDD4Fog supports the microservice architecture and provides the Test-Driven utilities such as metamorphic testing, mutation testing and random testing for the whole software development lifecycle of Fog Computing systems. To demonstrate the feasibility of TDD4Fog, we have presented some preliminary results on the key components of TDD4Fog and discussed some important future research directions.

Nowadays, autonomous driving has attracted much attention from both industry and academia. Convolutional neural network (CNN) is a key component in autonomous driving, which is also increasingly adopted in pervasive computing such as smartphones, wearable devices, and IoT networks. Prior work shows CNN-based classification models are vulnerable to adversarial attacks. However, it is uncertain to what extent regression models such as driving models are vulnerable to adversarial attacks, the effectiveness of existing defense techniques, and the defense implications for system and middleware builders.This paper presents an in-depth analysis of five adversarial attacks and four defense methods on three driving models. Experiments show that, similar to classification models, these models are still highly vulnerable to adversarial attacks. This poses a big security threat to autonomous driving and thus should be taken into account in practice. While these defense methods can effectively defend against different attacks, none of them are able to provide adequate protection against all five attacks. We derive several implications for system and middleware builders: (1) when adding a defense component against adversarial attacks, it is important to deploy multiple defense methods in tandem to achieve a good coverage of various attacks, (2) a black-box attack is much less effective compared with a white-box attack, implying that it is important to keep model details (e.g., model architecture, hyperparameters) confidential via model obfuscation, and (3) driving models with a complex architecture are preferred if computing resources permit as they are more resilient to adversarial attacks than simple models.

Privacy protection in electronic healthcare applications is an important consideration due to the sensitive nature of personal health data. Internet of Health Things (IoHT) networks have privacy requirements within a healthcare setting. However, these networks have unique challenges and security requirements (integrity, authentication, privacy and availability) must also be balanced with the need to maintain efficiency in order to conserve battery power, which can be a significant limitation in IoHT devices and networks. Data are usually transferred without undergoing filtering or optimization, and this traffic can overload sensors and cause rapid battery consumption when interacting with IoHT networks. This consequently poses restrictions on the practical implementation of these devices. As a solution to address the issues, this paper proposes a privacy-preserving two-tier data inference framework - this can conserve battery consumption by reducing the data size required to transmit through inferring the sensed data and can also protect the sensitive data from leakage to adversaries. Results from experimental evaluations on privacy show the validity of the proposed scheme as well as significant data savings without compromising the accuracy of the data transmission, which contributes to energy efficiency of IoHT sensor devices.

Advanced driver-assistance systems (ADASs) have been proposed as an alternative to driverless vehicles to provide support for automotive vehicle decisions. As a significant driving force for ADASs, the augmented reality (AR) provides comprehensive location-based content services for in-vehicle consumers. With the increase in request for information sharing, the current standalone mode of ADASs needs a shift to the multiuser sharing mode. In this article, to address the high mobility and real time requirements of ADASs in 5G environments, and also to address the resource orchestration and service management of big data in intelligent transportation systems, we integrate the information-centric network (ICN) and the network function virtualization (NFV) with ADASs to support an efficient AR-assisted content sharing and distribution. This integration eliminates the imbalance between the content requests and the resource limitation by splitting the virtual resources and providing an on-demand network and resource slicing in ADASs. We propose an incentive trading model for assistance content caching services and also propose a novel mechanism for optimal content cache allocation. Our extensive evaluation confirms that our proposed mechanism outperforms the past literature in terms of the cache hit ratio and latency.

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.

Android Apps have become the most important mobile applications in the evolving mobile IoT systems, whose security and privacy are confronted with ever more challenges, since such mobile devices as smartphones involve too much personal privacy information. Meanwhile, the developers prefer to put core functions (e.g., encryption function and T9 search function) of Android applications in the native layer for execution efficiency. However, there are no automated security analysis tools to protect the security and privacy of the Android native layer, especially for those dynamically loaded third-party SO libraries. In order to solve the previous problem, which is confusing, we propose a novel and scalable system, called SoProtector, to prevent privacy from leaking via the analysis of data flow between the Java and native layers. For detection of the malicious function implanted in the SO libraries, SoProtector realizes a real-time engine. We derive the malware features via three steps: 1) present binary files in native family as a grayscale image; 2) with use of the ARM instructions set reversely obtain the code of the SO file and using Python to obtain the opcode sequence; and 3) each file is transformed as the form of assembly language by IDA Pro, which includes a gdl file as an accompaniment. Our experiment, which involved 3400 applications, demonstrates that SoProtector is able to detect more sinks, sources, and smudges. It effectively inspects and blocks at least 82% of the applications that are loading malicious third-party SO dynamically, and it has relatively low overhead in the meantime, compared to most of the existing static analysis tools (e.g., FlowDroid and AndroidLeaks).

With the advancement of the Internet of Things (IoT) technologies, the number of heterogeneous IoT applications requiring a variety of resources and services is increasing dramatically. Recently, the introduction of fog computing has further unlocked the potential of real-time services within the IoT context. On the basis of fog architecture, we herein propose a novel rate-adaptive fog service platform aiming at heterogeneous services provisioning and optimized service rate allocation. By forming several service groups in the fog network in which each service could be adequately provisioned, service consumers would always benefit from the fact that the majority of services produced by the IoT applications are in their proximity and thus are delivered to the destination promptly. Taking advantage of the well-known network utility maximization (NUM) approach, a service rate-adaptive algorithm is developed to empower fog nodes working together to adjust service delivery rate dynamically. Throughout this process, the algorithm takes the current network condition and constraint into account to ensure the rate is calibrated in favor of providing satisfactory quality of service (QoS) to each service receiver at the same time. Compared to other resource allocation strategies that mainly focus on allocating resources for a single network service, our proposed platform is capable of not only dealing with both the elastic and inelastic services but also handling the abrupt network changes and converging back to the global optimum rapidly.

Outsourcing helps relocate data from the cyber-physical system (CPS) for efficient storage at low cost. Current server-based outsourcing mainly focuses on the benefits of servers. This cannot attract users well, as their security, efficiency, and economy are not guaranteed. To solve with this issue, a hybrid outsourcing model that exploits both cloud server and edge devices to store data is needed. Meanwhile, the requirements of security and efficiency are different under specific scenarios. There is a lack of a comprehensive solution that considers all of the above issues. In this work, we overcome the above issues by proposing the first hybrid user-centric data outsourcing (HUCDO) scheme. It allows users to outsource data securely, efficiently, and economically via different CPSs. Brielly, our contributions consist of theories, implementations, and evaluations. Our theories include the first homomorphic collision-resistant chameleon hash (HCCH) and homomorphic designated-receiver signcryption (HDRS). As implementations, we instantiate how to use our proposals to outsource small- or large-scale data through distinct CPS, respectively. Additionally, a blockchain with proof-of-discrete-logarithm (B-PoDL) is instantiated to help improve our performance. Last, as demonstrated by our evaluations, our proposals are secure, efficient, and economic for users to implement while outsourcing their data via CPSs.

Artificial Intelligence (AI) based technologies have deeply changed people's daily lives. There are many AI-based applications used in industrial scenarios such as Internet of Things (IoT), smart grids, and edge computing. Although bringing AI into industrial scenarios could improve the performance in many aspects, new security and privacy issues are also introduced consequently. Subsequently, machine learning technologies require a training process which introduces the protection problems in the training data and algorithms. As many machine learning and deep learning models are vulnerable against well-designed adversarial input samples, outsourcing data and algorithms for training will require the integrity of the training data. Also, data privacy of the end users must be protected. On the other hand, traditional solutions for industrial system security could also be enhanced by these AI schemes. The papers in this special section focus on emerging privacy and security issues brought by Artificial Intelligence in industrial informatics.

With the advent of 5G, the industrial Internet of Things has developed rapidly. The industrial sensor-cloud system (SCS) has also received widespread attention. In the future, a large number of integrated sensors that simultaneously collect multifeature data will be added to industrial SCS. However, the collected big data are not trustworthy due to the harsh environment of the sensor. If the data collected at the bottom networks are directly uploaded to the cloud for processing, the query and data mining results will be inaccurate, which will seriously affect the judgment and feedback of the cloud. The traditional method of relying on sensor nodes for data cleaning is insufficient to deal with big data, whereas edge computing provides a good solution. In this article, a new data cleaning method is proposed based on the mobile edge node during data collection. An angle-based outlier detection method is applied at the edge node to obtain the training data of the cleaning model, which is then established through support vector machine. Besides, online learning is adopted for model optimization. Experimental results show that multidimensional data cleaning based on mobile edge nodes improves the efficiency of data cleaning while maintaining data reliability and integrity, and greatly reduces the bandwidth and energy consumption of the industrial SCS.

Recommendation systems have been widely used in large e-commerce websites, but cold start and data sparsity seriously affect the accuracy of recommendation. To solve these problems, we propose SSL-SVD, which works to mine the sparse trust between users and improve the performance of the recommendation system. Specifically, we mine sparse trust relationships by decomposing trust impact into fine-grained factors and employing the Transductive Support Vector Machine algorithm to combine these factors. Then, we incorporate both social trust and sparse trust information into the SVD++ model, which can effectively utilize the explicit and implicit influence of trust for rating prediction in the recommendation system. Experiments show that our SSL-SVD increases the trust density degree of each dataset by more than 65% and improves the recommendation accuracy by up to 4.3%.

The Internet of Multimedia Things (IoMT) has become the backbone of innumerable multimedia applications in various fields. The wide application of IoMT not only makes our life convenient but also brings challenges to service discovery. Service discovery aims to leverage location information and trust evidence scattered in a variety of multimedia applications to find trusted IoMT devices that can provide specific service in target areas. However, the eavesdropping and tampering to these sensitive IoMT data during the trust propagation process invalidate the service discovery process. To address these challenges, we propose Secure Service Discovery (SSD) for IoMT using cross-blockchain-enabled fog computing. To resist the tampering and eavesdropping during the trust propagation process, a scalable cross-blockchain structure consisting of multiple parallel blockchains is first proposed based on fog, in which different parallel blockchains can be orchestrated to propagate encrypted location information and trust evidence of different applications. Moreover, to enable a cross-blockchain structure to leverage encrypted location information and trust evidence to find trusted IoMT devices in preset areas, a novel privacy-preserving range query is proposed to query and aggregate trust evidence. Security analysis and simulations are carried out to demonstrate the effectiveness and security of the proposed SSD.

Trust is widely applied in recommender systems to improve recommendation performance by alleviating well-known problems, such as cold start, data sparsity, and so on. However, trust data itself also faces sparse problems. To solve these problems, we propose a novel sparse trust recommendation model, SSL-STR. Specifically, we decompose the aspects influencing trust-building into finer-grained factors, and combine these factors to mine the implicit sparse trust relationships among users by employing the Transductive Support Vector Machine algorithm. Then we extend SVD++ model with social trust and sparse trust information for rating prediction in the recommendation system. Experiments show that our SSL-STR improves the recommendation accuracy by up to 4.3%.

Human activity recognition plays a key role in the application areas such as fitness tracking, healthcare and aged care support. However, inaccurate recognition results may cause an adverse effect on users or even an unpredictable accident. In order to improve the accuracy of human activity recognition, multi-device and deep learning based approaches have been proposed. However, they are not practical on a daily basis due to the limitations that devices are difficult to wear, and deep learning requires large training dataset and incurs expensive computational costs. To address this problem, we propose a novel approach, multiscale feature engineering for human activity recognition (MFE-HAR), which exploits the properties of arm movement from global and local scales using the accelerometer and gyroscope sensors on a single wearable device. Our method takes advantage of having important features at multiple scales over previous single-scale methods. We evaluated the performance of the proposed method on two public datasets and achieved the mean classification accuracy of 93% and 98% respectively. Our proposed system performs better than the state of the art multi-device based approaches, and is more practical for real-world applications.

The traditional methods of recognizing human activities involve typical machine learning (ML) algorithms which uses heuristic engineered features. Human activities are dynamic in nature and are encoded with a sequence of actions. ML methods are able to perform activity recognition tasks but may not exploit the temporal correlations of the input data. Therefore, in this paper, we proposed and showed the effectiveness of employing a new combination of deep learning (DL) methods for human activity recognition (HAR). DL methods are capable of extracting discriminative features automatically from the raw sensor data. Specifically, in this paper, we proposed a hybrid architecture which features a combination of Convolutional neural networks (CNN) and Long short-term Memory (LSTM) networks for HAR task. The model is tested on UCI HAR dataset which is a benchmark dataset and comprises of accelerometer and gyroscope data obtained from a smartphone. Our experimental results showed that our proposed method outperformed the recent results which used pure LSTM and bidirectional LSTM networks on the same dataset.

Summary Fog computing is used as a popular extension of cloud computing for a variety of emerging applications. To incorporate various design choices and customized policies in fog computing paradigm, Microservices is proposed as a new software architecture, which is easy to modify and quick to deploy fog applications because of its significant features, ie, fine granularity and loose coupling. Unfortunately, the Microservices architecture is vulnerable due to its wildly distributed interfaces that are easily attacked. However, the industry has not been fully aware of its security issues. In this paper, a survey of different security risks that pose a threat to the Microservices‐based fog applications is presented. Because a fog application based on Microservices architecture consists of numerous services and communication among services is frequent, we focus on the security issues that arise in services communication of Microservices in four aspects: containers, data, permission, and network. Containers are often used as the deployment and operational environment for Microservices. Data is communicated among services and is vital for every enterprise. Permission is the guarantee of services security. Network security is the foundation for secure communication. Finally, we propose an ideal solution for security issues in services communication of Microservices‐based fog applications.

The emergence of the mobile Internet of Things (IoT) has made our lives smarter, relying on its various mobile IoT devices and services provided. However, with the explosively emerging mobile IoT services, malicious attackers can access them in an unauthorized way. In this paper, we designed an Efficient Anonymous User Authentication (E-AUA) protocol between the users and servers based on multiserver architectures, which contain multiple servers to address the problem of network congestion in mobile IoT. Furthermore, the E-AUA protocol was designed with a dual messages mechanism with strong anti-attack ability, lower communication and computation costs. Comparing with the state of the art protocols, our E-AUA protocol reduced both communication and computation costs. We also provided a security analysis to demonstrate that our E-AUA protocol is secure and meets a variety of security requirements in a motivated mobile IoT scenario.

In recent years, wireless sensor networks (WSNs) have become an active area of research for monitoring physical and environmental conditions. Due to the interdependence of sensors, a functional anomaly in one sensor can cause a functional anomaly in another sensor, which can further lead to the malfunctioning of the entire sensor network. Existing research work has analysed faulty sensor anomalies but fails to show the effectiveness throughout the entire interdependent network system. In this article, a dictionary learning algorithm based on a non-negative constraint is developed, and a sparse representation anomaly node detection method for sensor networks is proposed based on the dictionary learning. Through experiment on a specific thermal power plant in China, we verify the robustness of our proposed method in detecting abnormal nodes against four state of the art approaches and proved our method is more robust. Furthermore, the experiments are conducted on the obtained abnormal nodes to prove the interdependence of multi-layer sensor networks and reveal the conditions and causes of a system crash.

Both academia and industry have directed tremendous interest toward the combination of Cyber Physical Systems and Cloud Computing, which enables a new breed of applications and services. However, due to the relative long distance between remote cloud and end nodes, Cloud Computing cannot provide effective and direct management for end nodes, which leads to security vulnerabilities. In this article, we first propose a novel trust evaluation mechanism using crowdsourcing and Intelligent Mobile Edge Computing. The mobile edge users with relatively strong computation and storage ability are exploited to provide direct management for end nodes. Through close access to end nodes, mobile edge users can obtain various information of the end nodes and determine whether the node is trustworthy. Then, two incentive mechanisms, i.e., Trustworthy Incentive and Quality-Aware Trustworthy Incentive Mechanisms, are proposed for motivating mobile edge users to conduct trust evaluation. The first one aims to motivate edge users to upload their real information about their capability and costs. The purpose of the second one is to motivate edge users to make trustworthy effort to conduct tasks and report results. Detailed theoretical analysis demonstrates the validity of Quality-Aware Trustworthy Incentive Mechanism from data trustfulness, effort trustfulness, and quality trustfulness, respectively. Extensive experiments are carried out to validate the proposed trust evaluation and incentive mechanisms. The results corroborate that the proposed mechanisms can efficiently stimulate mobile edge users to perform evaluation task and improve the accuracy of trust evaluation.

Network inference has been extensively studied to better understand the information diffusion in online social networks. In this field, state-of- art widely adopted a priori knowledge related to users' infection timestamps. Researchers also assume that the smaller the time difference between two nodes, the higher the likelihood of an edge between the pair of users. However, according to our technical analyses and empirical studies, existing methods have two critical problems 1) alternative spreading paths; 2) users' delivery delay, which leads to the inaccuracy of previous methods. In this paper, we developed an innovative method to address the inference inaccuracy caused by the exposed two problems. This method determined the existence of an edge between a pair of users according to part of the users' features. The experiment results suggested that our method achieved around 70% accuracy in inferring network structures while existing methods failed in the same tasks.

Among the approaches that investigate the similarity between web services, hardly any concentrates on the impacts from contexts. In this paper we introduce service2vec which is an approach to represent web services as service embeddings based on a recent popular deep learning technique word2vec. Our approach composes and combines web services to be a document that is trained by the modeling technique of word2vec. As a result, each web service in the document is vectorized. By taking the advantage of word2vec, the resulting service embeddings of service2vec can be used to illustrate the contextual relations between web services. The experimental results suggest that service2vec can deliver contextual similarity between web services.

Human activity recognition using embedded mobile and embedded sensors is becoming increasingly important. Scaling up from individuals to groups, that is, group activity recognition, has attracted significant attention recently. This paper proposes a model and specification language for group activities called GroupSense-L, and a novel architecture called GARSAaaS (GARSA-as-a-Service) to provide services for mobile Group Activity Recognition and Situation Analysis (GARSA) applications. We implemented and evaluated GARSAaaS which is an extension of a framework called GroupSense where sensor data, collected using smartphone sensors, smartwatch sensors and embedded sensors, are aggregated via a protocol for these different devices to share information, as required for GARSA. We illustrate our approach via a scenario for providing services for tour leaders aiding Vehicle-to-Human (V2H), Vehicle-to-Group (V2G) and Vehicle-to-Vehicle (V2V) interactions to increase the group safety. We demonstrate the feasibility of our model and expressiveness of our proposed model.

In Cyber-Physical Systems (CPS), cyber and physical components must work seamlessly in tandem. Runtime verification of CPS is essential yet very difficult, due to deployment environments that are expensive, dangerous, or simply impossible to use for verification tasks. A key enabling factor of runtime verification of CPS is the ability to integrate real-time simulations of portions of the CPS into live running systems. We propose a verification approach that allows CPS application developers to opportunistically leverage real-time simulation to support runtime verification. Our approach, termed B race B ind , allows selecting, at runtime, between actual physical processes or simulations of them to support a running CPS application. To build B race B ind , we create a real-time simulation architecture to generate and manage multiple real-time simulation environments based on existing simulation models in a manner that ensures sufficient accuracy for verifying a CPS application. Specifically, B race B ind aims to both improve simulation speed and minimize latency, thereby making it feasible to integrate simulations of physical processes into the running CPS application. B race B ind then integrates this real-time simulation architecture with an existing runtime verification approach that has low computational overhead and high accuracy. This integration uses an aspect-oriented adapter architecture that connects the variables in the cyber portion of the CPS application with either sensors and actuators in the physical world or the automatically generated real-time simulation. Our experimental results show that, with a negligible performance penalty, our approach is both efficient and effective in detecting program errors that are otherwise only detectable in a physical deployment.

Cyber-Physical Systems (CPS) have gained wide popularity, however, developing and debugging CPS remain significant challenges. Many bugs are detectable only at runtime under deployment conditions that may be unpredictable or at least unexpected at development time. The current state of the practice of debugging CPS is generally ad hoc, involving trial and error in a real deployment. For increased rigor, it is appealing to bring formal methods to CPS verification. However developers often eschew formal approaches due to complexity and lack of efficiency. This paper presents Brace Assertion, a specification framework based on natural language queries that are automatically converted to a determinitic class of timed automata used for runtime monitoring. To reduce runtime overhead and support properties that reference predicate logic, we use a second monitor automaton to create filtered traces on which to run the analysis using the specification monitor. We evaluate the Brace Assertion framework using a real CPS case study and show that the framework is able to minimize runtime overhead with an increasing number of monitors.