Federated learning is a collaborative training paradigm designed to protect private data and is widely used in the cooperative training of Internet of Things (IoT) devices. However, despite its focus on privacy protection, federated learning remains susceptible to poisoning attacks from malicious clients. These attacks can degrade system performance and potentially lead to data privacy breaches. Moreover, real-world IoT datasets are often heterogeneous, further increasing the difficulty of detecting malicious clients. Existing defense mechanisms often struggle to effectively identify malicious clients while maintaining high model performance. To address this issue, we propose a defense mechanism called Malicious client exclusion aggregation (MaEA). This method utilizes KL divergence to preliminarily filter out anomalous clients, aggregates the remaining (preliminarily filtered) clients to obtain a pre-center model, and then identifies and excludes malicious clients by measuring their deviations from this pre-center model. We executed a series of extensive experiments on the CIFAR-10 dataset to demonstrate the effectiveness of MaEA. The results demonstrate that our approach can efficiently detect and identify malicious clients while correcting model performance.

Smart homes depend on collaborative sequential prediction tasks to optimize energy consumption and appliance scheduling. Federated learning (FL) offers a promising approach by enabling decentralized model training to balance privacy and usability. Yet, standard FL techniques fail to effectively address data diversity and individual user preferences in smart home contexts. To address these issues, we propose P3FL: a Privacy-Preserving Personalized Federated Learning framework that integrates tailored model training and privacy enhancements for federated collaborative predictions and decision-making. Our framework introduces the Personalized Collaborative Decision-Making (PCDM) algorithm, which dynamically adapts to different household environments while ensuring privacy and personalization. P3FL combines a global model for knowledge aggregation with a personalized adaptation module to provide fine-tuned predictions based on user preferences, environmental factors, and device configurations. Theoretical convergence bounds analysis confirms the robustness and efficiency of PCDM under conditions of strong convexity, smoothness, and bounded variance. Extensive experiments on real-world smart home datasets demonstrate that P3FL outperforms state-of-the-art methods, with PCDM achieving a training accuracy of 92.14%. Our approach enhances operational efficiency and ensures personalized user satisfaction, privacy enhancement in smart homes.

Eigenvalue decomposition (EVD) is a fundamental yet time-consuming operation with extensive applications in Internet of Things (IoT). When the matrix dimension reaches millions, resource-limited IoT devices struggle to perform such computationally expensive operations. Edge computing, with its plentiful computing resources, offers an effective solution to this problem. However, privacy concerns arise because outsourced tasks may contain sensitive user data. In this article, we propose the first privacy-preserving, edge-assisted EVD outsourcing scheme that securely enables users to outsource EVD tasks to edge servers. We design a privacy-preserving matrix transformation method to encode the original data, ensuring that edge servers cannot access users’ private information. Additionally, we design a verification scheme that enables the user to verify the correctness of the results returned by the edge servers. Our protocol supports parallel computation by multiple edge servers, thus enhancing the efficiency of EVD. The feasibility of our proposed scheme is demonstrated through both theoretical and experimental perspectives.

Embodied AI has notably advanced the autonomy of physical agents such as robots, vehicles, and AAVs, expanding their application scope. However, existing systems are predominantly data-driven, relying on static programming and pre-trained models. This limits their adaptability to dynamic and unforeseen scenarios. Additionally, the high computational cost of training large-scale models locally hinders their practical deployment. One promising solution lies in integrating Large Language Models (LLMs) into Embodied AI frameworks. Although LLMs excel in reasoning, coding, and perception, most existing frameworks adopt a single-LLM architecture, which restricts their effectiveness in addressing complex, multimodal tasks. The diverse strengths of individual LLMs, ranging from natural language understanding, visual processing to code generation, are seldom utilized in a collaborative and structured manner. To address these challenges, we propose a knowledge-driven framework, called EUF, that incorporates multi-LLMs into the Embodied AI architecture for AAV-as-a-Service in Intelligent Transportation Systems. Each LLM is dedicated to a specific stage of the AAV task, including user intent interpretation, adaptive path planning with code generation, and error-feedback mechanisms. Our research explores both One-shot and Segmented Code Generation approaches using various LLM-driven models to identify the optimal strategy. We conduct an in-depth analysis of different code errors to evaluate the strengths and limitations of each approach, including the feedback capabilities of the LLM-driven models. Experiments conducted in the AirSim environment demonstrate the framework’s robustness and accuracy in complex AAV path planning tasks, highlighting its practical potential for real-world ITS deployments.

In recent years, the rapid development of the Industrial Internet of Things (IIoT) has enabled real-time communication and data sharing among devices, significantly enhancing industrial production efficiency and security. Furthermore, the introduction of edge learning allows models to be trained on edge industrial devices. However, with the continuous growth of industrial data, challenges such as resource optimization in edge environments, edge node motivation, and threats from malicious nodes are increasingly posing obstacles to the advancement of edge learning. In this paper, we propose Blockchain and Coded Computing based Secure Edge Learning (BCC-SEL). First, we introduce a coded edge learning framework with Lagrange Coded Computing (LCC) for resource-efficient use of idle nodes during training. Based on the blockchain, we further propose an incentive mechanism to reward and punish the participating training clients. Finally, we guarantee the robustness of the framework using the detection method based on cosine-similarity. We provide theoretical proof that our approach effectively reduces the computational consumption of training nodes. In terms of experiments, our method effectively rewards honest nodes that participate in training and penalizes malicious nodes while guaranteeing accuracy.

Singular Value Decomposition (SVD) is a widely applied foundational decomposition technique; however, its computational demands often exceed the capabilities of Internet of Things (IoT) devices. While leveraging edge servers can alleviate this load, it may introduce potential security vulnerabilities. Current secure outsourcing computation methods designed for cloud environments are challenging to adapt to distributed schemes in edge computing. Our research proposes a novel secure edge-assisted protocol for IoT devices solving SVD, aiming to conceal the Input/Output matrix and balance computational loads across multiple edge servers. The protocol ensures the confidentiality of original matrices and decomposition results, preventing exposure to edge servers. We conduct a comprehensive theoretical analysis of the protocol’s efficiency and security, substantiating its advancements through experiments.

The technology of the Internet of Vehicles (IoV) and digital twins (DTs) is driving deeper connectivity between vehicles and road infrastructure. Through the data exchange of IoV and the simulation of DT technology, vehicle driving decisions, traffic management, and road planning are optimized. However, DT models contain a large amount of private vehicle data, causing the risk of privacy leakage. Distributed artificial intelligence (AI) methods, particularly federated learning (FL) algorithms, ensure data security and privacy by sharing data models rather than sharing private data. Current mainstream algorithms use FL and local differential privacy (LDP) or blockchain approaches to protect data security at the cost of lower model accuracy and larger computation time. In the vehicle road cooperation, we designed a three-layer DT-driven personalized privacy-preserving framework, which includes a physical layer, a DT layer, and an application layer. In our proposed framework, to improve the security and performance of DT models, a time-sensitive PLDP-based FL (TimeSenFLDP) mechanism is proposed to achieve different privacy levels of the DT model of vehicles over sharing time steps. Compared with the mainstream algorithm (e.g., DP-SGD), the experiments prove that our proposed algorithm has 18.07%, 16.32%, and 7.5% accuracy improvement in FedAvg, FedProx, and FedDyn, respectively.

The Internet of Vehicles (IoV) connects a massive amount of smart vehicles for inter/intra-vehicle information sharing. Data privacy issues, such as privacy leakage and privacy cost are the key challenges that hinder vehicle operators from sharing their data safely. Traditional privacy-preserving techniques, including Federated Learning (FL) and Differential Privacy (DP) techniques, can protect data privacy and security, but the high privacy cost severely limits learning performance. In addition, the IoV services place high demands on low communication latency, which can be obtained by reducing the communication bits, but it also limits the learning performance. Thus, how to solve the communication-privacy-accuracy tradeoffs to achieve low latency, high privacy preservation and model performance has been a complicated issue in IoV. In this paper, a privacy-enhancement differentially private federated learning framework (FedSDP) is proposed based on the shuffle model to ensure secure and efficient data sharing under the constraint of low latency in IoV. In our proposed framework, four privacy enhancement methods are proposed, including data subsampling, vehicle sampling, shuffle model and dummy points, to amplify the privacy and obtain higher learning performance. Then, a Top-K sparsification mechanism of the vehicle training process is proposed to reduce communication bits. Finally, the experimental results indicate that our approach can reduce the communication latency by 31.66%, enhance the privacy ϵc by 30.77% and improve the test accuracy by 48.56%, compared with the traditional SDP mechanism.

Blockchain data mining has the potential to reveal the operational status and behavioral patterns of anonymous participants in blockchain systems, thus providing valuable insights into system operation and participant behavior. However, traditional blockchain analysis methods suffer from the problems of being unable to handle the data due to its large volume and complex structure. With powerful computing and analysis capabilities, graph learning can solve the current problems through handling each node's features and linkage relationships separately and exploring the implicit properties of data from a graph perspective. This paper systematically reviews the blockchain data mining tasks based on graph learning approaches. First, we investigate the blockchain data acquisition method, integrate the currently available data analysis tools, and divide the sampling method into rule-based and cluster-based techniques. Second, we classify the graph construction into transaction-based blockchain and account-based methods, and comprehensively analyze the existing blockchain feature extraction methods. Third, we compare the existing graph learning algorithms on blockchain and classify them into traditional machine learning-based, graph representation-based, and graph deep learning-based methods. Finally, we propose future research directions and open issues which are promising to address.

The Internet of Vehicles (IoV) based on the 6-th Generation (6G) communication brings convenience, but also raises anxiety about information security. Researchers have developed static security schemes based on the blockchain, but it results in excessive resource occupation. And it is difficult to resist some attacks such as desynchronization, Denial of Service (DoS), and covert intrusion. In response to the above problems, this paper proposes the Floating Blockchain Consensus Security (FBCS) scheme. It models the attack risk of Road Side Unit (RSU) through the traffic flow prediction to present the credibility evaluation and constructs the floating blockchain based on the results. And the security capability is adjusted through the dynamic joining and exit mode of trusted nodes and untrusted nodes. FBCS also establishes a relationship between blockchain size and energy consumption. Once the system resources are found to be insufficient, it applies the cloud center-based supplementary certification mechanism to provide authentication to untrusted nodes to enhance the stability of the IoV.Theoretical analysis and simulation experiments prove that the FBCS can afford data privacy, maintain moderate security capabilities, and adapt defense capabilities according to the attack environment, reduce resource occupation to save cost.

The industrial Internet of Things (IoT) system is integrated with the emerging artificial intelligence (AI) paradigms to empower industrial automation and self-evolving capabilities. AI-driven resource allocation across cyber-physical domains for mobile industrial IoT must consider its fundamental requirements and key characteristics such as high reliability, low latency, and environmental dynamics. The challenge is twofold. Industrial systems are fault-sensitive, which makes them intolerable of trial-and-error-based learning and optimization approaches. In addition, learning models cannot adapt to changing industrial IoT environment with dynamic communication noise and machinery disturbances. In this paper, we propose joint optimization for the nonorthogonal multiple access (NOMA) and multi-tier hybrid cloud-edge computing empowered industrial IoT that results in improved utilization of communication and computing resources. Second, we establish the fine-grained digital twin for industrial IoT (DT-IIoT) to simulate the changing industrial environment to support trial-and-error-based safe learning. Third, we leverage meta reinforcement learning (meta RL) to improve the generalization and fast adaptation of the learning models for DT-IIoT. Finally, the feasibility and efficiency of these schemes are evaluated through extensive experiments.

The industrial visual sensing-actuation system is an implementation approach to construct the loop between the digital twin and physical systems, which is facing the following challenges. First, the cross-digital-physical information exchanges bring a high end-to-end delay that threatens the functional safety of industrial systems. Second, industrial scenarios are diverse, such as manufacturing, chemical engineering, etc., which makes the intelligent sensing strategies for one scenario inapplicable to others, especially for few-shot cases. Third, intelligent actuation strategies cannot allocate resources across digital and physical domains. We propose the delay-minimization-based intelligent digital twin approach to address the above challenges. The digital twin framework incorporates samples from the physical domain to train the learning models in the digital domain. The proposed scheme tailors and adapts transferable and reinforced learning models with end-to-end delay analysis to optimize the training process. The feasibility and efficiency of the scheme are validated by simulations.

Advanced persistent threats (APT) have novel features such as long-term latency, precision strikes and uncertain strategies. APT poses severe threats to the resource-limited edge devices in advanced networks. Cyber threat intelligence (CTI) conducts data analysis on attack strategies by artificial intelligence (AI) and generates threat intelligence to optimize the detection model and guide defense strategies. However, AI lacks explanations for the decisions and thus reduces the transparency and performance of the detection model. Besides, the tradeoff between the detection accuracy and the computational resource limitation of edge devices needs an optimal and rapid dynamic resource allocation method, which edge game and AI can help. In this paper, we propose an explainable intelligence-driven APT edge defense mechanism. The proposed mechanism provides guidelines and explanations for designing the defense strategy and resource allocation scheme of the edge defender to detect APT. The edge defense strategy model is based on edge Bayesian Stackelberg game and CTI. Meanwhile, we implement a DRL-based resource allocation scheme to meet rapid response requirements at the edges. We demonstrate that the proposed mechanism can improve the protection level of edges and defense capability against APT through extensive experiments.

Industrial Internet of Things (IIoT) envisions the tight coupling of numerous critical industrial manufacturing subsystems, such as control, networking, and computing through the ubiquitous Internet of Things technologies. Nonetheless, such interconnectivity poses significant challenges to the successful management and operation of massively distributed industrial manufacturing systems. Without carefully integrated system design, the nonoptimal management and operation of highly intertwined subsystems can lead to the loss of productivity and ultimately the value of factories and plants. To address this issue, in this article, we conduct the integrated design that is capable of simultaneously configuring both control and networking subsystems in IIoT with consideration for their inherent interdependencies. We first analyze the performance of the dynamic backoff exponential (BE) in IEEE 802.15.4 carrier-sense multiple access (CSMA) and show the performance improvement of dynamic BE. We then design a model-free reinforcement learning algorithm to configure the control and networking subsystems automatically via systematic trial and error, as it is impractical to build a model for a highly intertwined complex IIoT system. Considering the time-sensitive characteristics of IIoT systems, we design priority-aware policies based on importance among networking traffic (i.e., sensing traffic and actuation traffic) to improve the convergence speed. The experimental results demonstrate that our priority-aware reinforcement-learning-based integrated design can successfully reconfigure the complex and highly intertwined IIoT system at runtime with a minimal convergence time. Besides, our approach reduces convergence time by 37.5%, and energy consumption by 9.2%, compared to the standard reinforcement learning approach.

The cybertwin and 6G-enabled Industrial Internet of Things (6G-IIoT) are the critical technologies that create the digital counterparts for physical systems and enable the near-instant interconnectivity in the industrial domain. It is in demand but challenging to conduct the integrated design for 6G-IIoT, which intertwines the cyber subsystems, such as control, communication, computing (3C), and the physical industrial factories and plants. Therefore, the cybertwin, which synchronizes between the digital counterparts and its physical entities during the system runtime, is the ideal proving ground for conducting the integrated design on the highly intertwined 3C of 6G-IIoT. However, the cybertwin lacks artificial intelligence to capacitate the automated integrated design for the 6G-IIoT. In this article, we first demonstrate the architecture of the machine-learning-based cybertwin for 6G-IIoT. Then, we leverage deep reinforcement learning (DRL) to conduct the integrated design via systematic trial and error in the cybertwin model, which is otherwise costly and dangerous in real industrial systems. Moreover, we invent the adaptive observation window for deep$Q$-network (AOW-DQN), which generates system states adaptive to the control system’s physical dynamics. Finally, the experimental results demonstrate the effectiveness and efficiency of our approach. To the best of our knowledge, we are the first to present the machine-learning-based cybertwin for carrying out the integrated design on the 3C for 6G-IIoT.

Industrial Internet-of-Things (IIoT), also known as Industry 4.0, is the integration of Internet of Things (IoT) technology into the industrial manufacturing system so that the connectivity, efficiency, and intelligence of factories and plants can be improved. From a cyber physical system (CPS) perspective, multiple systems (e.g., control, networking and computing systems) are synthesized into IIoT systems interactively to achieve the operator's design goals. The interactions among different systems is a non-negligible factor that affects the IIoT design and requirements, such as automation, especially under dynamic industrial operations. In this paper, we leverage reinforcement learning techniques to automatically configure the control and networking systems under a dynamic industrial environment. We design three new policies based on the characteristics of industrial systems so that the reinforcement learning can converge rapidly. We implement and integrate the reinforcement learning-based co-design approach on a realistic wireless cyber-physical simulator to conduct extensive experiments. Our experimental results demonstrate that our approach can effectively and quickly reconfigure the control and networking systems automatically in a dynamic industrial environment.

Unmanned Aerial Vehicles (UAVs), also known as drones, have become a new paradigm to provide emergency wireless communication infrastructure when conventional base stations are damaged or unavailable. In this paper, we propose new schemes to enable the 3D deployment of drones, which can provide network coverage and connectivity services for users located in uneven terrain. We formalize two models, including optimal coverage model and optimal connectivity model, which belong to NP-hard. To be specific, we first consider both the quality of service (QoS) requirements of users and the capacity of drones. We then formalize the problem and design a heuristic scheme, called Particle Swarm Optimization (PSO) algorithm to achieve a cost-effective solution. We also address the optimal connectivity problem in a scenario, in which a number of isolated local networks have been established by users through ad hoc communication and/or device-to-device (D2D) communication. We further develop the cost-effective heuristic algorithm to effectively minimize the total number of required drones. Via extensive performance evaluation, our experimental results demonstrate that the proposed schemes can achieve the effective deployment of drones for users in uneven terrain with respect to the number of required drones.

Within the foreseeable future, the growing number of mobile devices, and their diversity, will challenge the current network architecture. Furthermore, users will expect greater data rates, lower latency, lower packet drop rates, etc. in future wireless networks. Ultra Dense Networks (UDN), considered to be one of the best ways to meet user expectations and support future wireless network deployment, will face multiple significant hurdles, including interference, mobility, and cost. In this paper, we review existing research efforts toward addressing those challenges and present future avenues for research. We first develop a taxonomy to review and describe existing research efforts. Next, we focus on inter-cell interference, handover performance, and energy efficiency as the key techniques to addressing the most pressing challenges. Finally, we present several future research directions, including emergent Internet-of-Things (IoT) applications, security and privacy, modeling and realistic simulations, and relevant techniques.

The Ultra Dense Network (UDN), as a key enabler for future wireless networks (such as 5G), is comprised of a massive number of small cells in the network. Nonetheless, energy consumption will be non-negligible when a large number of smallcell Base Stations (BSs) are densely deployed. One practical and effective approach to reduce the energy consumption of the UDN is through dynamically controlling the power saving mode of BSs, while the challenge is to maintain network coverage and satisfy the performance requirements of User Equipment (UEs). In this paper, we formalize the problem of minimizing the energy consumption of BSs by optimally controlling the BS's power saving mode (switching between awake mode and sleep mode). We focus on optimal BS selection with the objective of energy efficiency, while the considering the constraints of the coverage of UEs, the capacity of BSs, and the data rate UEs. To validate the effectiveness of our proposed scheme, we have conducted performance evaluations with a comprehensive scenario design, consisting of UE density, distribution, and mobility, as well as BS deployment. The evaluation results demonstrate favorable energy efficiency improvement at averages of 38.82 % and 48.05 % in scenarios where UEs are uniformly distributed and non-uniformly distributed in the network. Meanwhile, network coverage and UE's Quality of Service (QoS) requirements are provided.

Information security is key important when we are trying to interconnect the wireless body sensor network with the healthcare social network via mobile facilities. In this paper, we specially work on a secured electrocardiogram (ECG) signal transmission scheme to prevent further injuries for patients with heart diseases from human emotional stress. We proposed a dynamic encryption method via biometric information among frequency spectrums of ECG signals, which can guarantee both high classification rate (>90 %) and system energy efficiency. At the same time, cooperative relays are applied for an additional spatial diversity gains. Simulation results show that the improved transmission rate and signal power capacity can lower the probability of data intercept (LPI) and detection (LPD) by taking the advantages of both temporal and spatial diversities. The network security thereby can be further improved.

In this work, we try to collect useful emotional stress information from electrocardiogram (ECG) signals via a real-time wearable Wireless Body Area Network (WBAN). Discrete Wavelet Transform (DWT) is applied on collected ECG signals for feature extraction, which carries important information for stress level identification. After the stress level is classified using K-Nearest Neighboring (KNN), adaptive convolutional coding is considered for ECG signal protection during transmission according to their various stress levels, which is able to provide an acceptable low Bit Error Rate (BER) and efficient energy consumption at the same time.