Self-supervised anomaly detection has emerged as a research hotspot in intelligent manufacturing and quality inspection, holding significant practical value in industrial applications. However, anomaly detection in real-world printed circuit board (PCB) production environments remains challenging. Existing methods often exhibit limited generalization when facing diverse anomaly types and environmental disturbances. In addition, high model complexity and insufficient capability for multi-scale fine-grained defect recognition constrain their practical deployment. To address these issues, this paper proposes a novel self-supervised anomaly detection framework (FSDNet). First, this paper proposes an Anomalous Sample Synthesizer Based on Diffusion Model (AnoDiff), which generates diverse and controllable anomalous samples to improve model generalization. Second, this paper designs an Anomaly Feature Perception Module (AFPM) that selects discriminative channels from pretrained features, thereby reducing model complexity while enhancing detection performance. Third, this paper proposes a Multi-scale Residual Reconstruction Network (MRRN) is developed to aggregate multi-scale features, improving sensitivity to fine-grained anomalies. Finally, this paper proposes two novel attention-based modules: a Top-k Sparse & Space Attention Module (TSSM) and a Gated Feature Enhancement Module (GFEM), both of which strengthen the discriminability and robustness of anomaly features. Experimental results demonstrate that the proposed method significantly outperforms state-of-the-art approaches on MVTec-AD, ViSA, and a self-constructed PCB dataset in terms of detection accuracy and robustness, validating its effectiveness and practical utility. The dataset and code are available at https://github.com/QinLi-STUDY/FSDNet/tree/master .

As an important part of mechanical equipment, rolling bearing holds significant importance in the normal operation of machinery. However, the parameter and computation of fault diagnosis approaches based on deep learning technique are huge, and most methods are diagnosed in the cloud, which can lead to time delays and non-real-time. To overcome these issues, in edge computing scenarios, a real-time bearing fault diagnosis network MobiKanViT is proposed in this paper. The network is a lightweight and low-latency vision fault diagnosis network with enhanced discriminative feature learning capability. The Efficient Multi-scale Attention (EMA) module is introduced to enhance the ability of feature recognition. The ordinary convolution module is substituted with the Kolmogorov-Arnold Networks (KAN) convolution module to solve the problem of large parameter and computation. The MobiKanViT is compressed and quantized with depth parameter γ to make the model more lightweight and easy to be deployed on edge devices. Verification experiments were conducted on two sets of experimental equipment, and three mobile phones were selected as mobile edge computing platforms. The experimental results show that a depth parameter of γ = 0.5 and INT8 quantization yield the most effective results for MobiKanViT. When juxtaposed with current large model techniques, the suggested approach decreases memory consumption by an average of 94.6%, while simultaneously boosting inference speed by about 15.87 times. In comparison to existing lightweight models, this new method also improves diagnostic accuracy by an average of 2.6%.

Compound faults, which arise from the interaction of multiple simultaneous failures, pose significant risks to the maintenance of industrial machinery in the Industrial Internet of Things (IIoT), leading to complex and unpredictable system failures. Traditional models tend to misclassify emerging compound faults as known categories due to a bias toward seen data. Additionally, the delayed emergence of compound faults relative to single faults hampers prompt sample collection. Furthermore, compound fault datasets typically exhibit a long-tailed distribution. Driven by these challenges, we propose an incremental contrastive learning model based on cross-modal contrastive embedding (ICLCFD) to achieve an incremental fault diagnosis from single faults to compound faults.Firstly, we employ Symmetrized Dot Pattern (SDP) image transformation to convert vibration signals into visual representations. We extract high-dimensional visual features from these SDP images using a Multi-Scale Residual Convolutional Neural Network (MS-ResCNN) and generate low-dimensional semantic features based on vibration signals to obtain richer feature information. Subsequently, a novelty detection mechanism is developed using the predicted number of fault sources as a count-based indicator to identify newly emerging faults. Furthermore, we incorporate a difficulty-aware class rebalancing sampling strategy to prioritize hard-to-diagnose samples during incremental updates, mitigating the excessive impact of head-class samples on the diagnosis results. Experiments on three open-source bearing datasets (CWRU, PU, and XJTU-SY) demonstrate that ICLCFD achieves a state-of-the-art accuracy of 96.58 ± 0.28%, outperforming existing methods by approximately 2%. The results confirm its effectiveness in handling unknown and compound faults in dynamic IoT maintenance pipelines.

Underwater acoustic backscatter communication technology brings a new perspective on addressing the energy dilemmas of Internet of Underwater Things (IoUT). However, time asynchronism in underwater acoustic backscatter devices (UABDs) can significantly degrade the performance of the UABD-based IoUT system. Existing time synchronization algorithms lose practicability leading to high energy consumption in scenarios where charging delays vary. To address these challenges, we propose UAB-Sync, a time synchronization algorithm specifically designed for the system. UAB-Sync introduces a novel three-stage architecture that uses dual constraints of time and energy to dynamically optimize the duration of energy signal, achieving adaptive approximation of optimal results. Besides, a closed-form solution for clock parameter estimation that incorporates Doppler factor estimation and accounts for multi-source measurement errors is developed, ensuring effective synchronous correction. Simulation results demonstrate that UAB-Sync significantly outperforms existing synchronization schemes in terms of both accuracy and energy efficiency for the UABD-based IoUT system.

As low Earth orbit (LEO) satellite constellations expand, the volume of on-orbit data processing increases significantly. However, these systems face challenges in data processing due to heterogeneous data distribution and limited onboard resources. This paper presents an innovative framework for personalized federated learning (PFL) tailored to heterogeneous LEO satellite networks, which mitigates data processing challenges and optimizes distributed computational performance across the satellite constellation. Our approach introduces personalized models built on individual satellite datasets, coupled with dynamic model aggregation and pruning techniques for efficient training. By overcoming data heterogeneity and localizing global models, our method significantly improves performance over traditional approaches. Extensive simulation validates the effectiveness of our PFL framework, which demonstrates its superiority in integrating FL with model pruning in satellite networks. Simulation results show that our proposed PFL method outperforms four comparative algorithms, which highlights its effectiveness in addressing the unique challenges of LEO satellite networks.

In this paper, we propose a novel cooperative relay-based resource and learning optimization (CRRLO) scheme that extends device connectivity, balances learning contributions, and coordinates communication resources to mitigate the negative impact of intermittent participation on FL performance caused by unreliable communication in Industrial Internet of Things (IIoT) environments. After local training, a cooperative aggregation stage is proposed, where fully connected device-to-device (D2D) relaying enables devices with failed device-to-server (D2S) transmissions to still contribute to the global model, while avoiding the transmission burden and relay selection issues associated with single-relay strategies. To further ensure unbiased aggregation, we produce reliability-driven aggregation weights to calibrate each device’s contribution to the global update. We then formulate a joint optimization problem aimed at improving FL convergence rate under communication and resource constraints by co-optimizing blocklength, transmission power, and aggregation weights. An iterative algorithm is designed to determine blocklength bounds, a low-complexity method is developed for power optimization, and a convex relaxation approach is adopted for weight adjustment. These subproblems are alternately solved using a block coordinate descent (BCD) method. Simulation results demonstrate that the proposed CRRLO scheme significantly accelerates convergence and improves test accuracy by up to 24.33% compared to baseline schemes under high transmission error probability.

The accuracy of digital twin models hinges on the prompt collection of information from the vehicular environment. However, the high mobility of vehicles and the dynamically changing network environment pose significant challenges. Dynamic twin model migration can reduce the Age of Information (AoI) by bringing twin models closer to their vehicles. Existing works rarely consider the inherent differences in optimization cycles between digital twin model migration and data upload, which potentially leads to suboptimal cost efficiency and information freshness. Specifically, real-time vehicular data must be rapidly uploaded to edge servers to ensure the accuracy and timeliness of digital twin models, while frequent migration of twin models over short periods incurs substantial costs. Therefore, we propose a dual-timescale bilevel learning approach, where the upper-layer learning optimizes twin model migration decisions on a long timescale to achieve forward-looking model migration, and the lower-layer learning optimizes data upload and resource allocation decisions on a short timescale to ensure the accuracy and timeliness of digital twin models. Then, we design a multi-agent selective parameter sharing approach based on spatiotemporal dependency correlations to accelerate model convergence and reduce communication costs among agents. Furthermore, through a rigorous theoretical analysis, we prove the convergence of the dual-timescale bilevel learning with broad applicability. Finally, numerical results demonstrate that our algorithm outperforms comparison algorithms in terms of convergence, AoI, and system cost, achieving at least a 21.30% reduction in AoI and a 14.58% reduction in system cost compared to the benchmark algorithms.

In mobile edge computing (MEC), computation offloading serves as an effective solution to bridge the gap between the stringent latency requirements of computational tasks and the limited processing capabilities of terminal devices (TDs). However, complex inter-task dependencies, dynamic network conditions, and the decentralized architecture of MEC systems pose significant challenges to efficient and adaptive task offloading. To address these challenges, this paper investigates dependency-aware task offloading and resource optimization in MEC environments. First, we propose a task feature extraction method based on dependency-aware graph neural networks (FEDG), which captures the hierarchical structure and varying importance of subtask dependencies by adaptively learning the aggregation weights of predecessor nodes and edges. Then, to address the joint dependency-aware task offloading and resource allocation problem under partial observability in MEC networks, we design a Dependency-aware Graph-based Multi-Agent deep reinforcement learning (DGMA) algorithm. DGMA integrates adaptive prioritized experience replay and correlation-based selective parameter sharing to improve learning efficiency and accelerate convergence in multi-agent environments. Extensive simulations demonstrate that DGMA achieves superior performance in terms of delay, energy consumption, offloading utility, and deadline violation rate.

Supported by space-air-ground integrated networks (SAGIN), Internet of Remote Things (IoRT) is regarded as a cornerstone for realizing global connectivity in 6 G networks. The integration of high-altitude platforms (HAPs) and unmanned aerial vehicles (UAVs), offering both wide coverage and agile data access, becomes a promising paradigm for IoRT data collection. However, sustaining reliable and efficient transmission is challenged by the mobility and constrained onboard energy of HAPs and UAVs, as well as atmospheric fading effects. To address these issues, we propose a green and efficient HAP-UAV collaborative design for IoRT data collection, which jointly considers both transmission performance and energy consumption. Firstly, we introduce a novel metric, Overall Energy Efficiency (OEE), to quantify the balance between cooperative transmission performance and the total energy cost under dynamic trajectory planning. Secondly, we formulate a joint optimization problem that simultaneously optimizes UAV/HAP trajectories, UAV power control, HAP selection, and bandwidth allocation. Thirdly, to address the formulated non-convex fractional problem, we develop an energy efficiency maximization strategy based on the successive convex approximation technique. Extensive simulation results demonstrate that the proposed strategy achieves significant gains in OEE, achieving superior trade-offs between energy consumption and transmission performance in HAP-UAV-assisted IoRT networks.

With the rapid development of vehicle-to-everything (V2X) technology, service migration has become an important approach to provide low-latency computing services and ensure service continuity for high-speed moving vehicles in vehicular edge computing (VEC), which enables VEC to efficiently support advanced transportation services. However, optimizing service satisfaction for service migration in multi-vehicle heterogeneous VEC networks is challenging, since the complex, multifactorial, and nonlinear dependencies between service satisfaction and quality of service (QoS) metrics is intractable, and the rapidly changing computational loads in edge server results in inefficient utilization of edge resources. In this paper, we propose a service Satisfaction-based Adaptive service Migration and resource Allocation joint Optimization scheme (SAMAO) to improve service migration efficiency and edge resource utilization in VEC. Firstly, we develop an adaptive computation resource allocation algorithm that can adjust resource allocation strategy according to load status of edge servers to improve vehicle service satisfaction. Then, to minimize energy consumption and ensure service satisfaction for vehicles, we propose a utility maximization algorithm to formulate migration decisions based on pre-allocated computation resources on servers. Finally, numerous simulations based on Shanghai Telecom real-world dataset show that SAMAO can achieve significant advantages in terms of average service satisfaction and computation cost.

In intelligent transportation systems, the ability to make precise and efficient lane-changing overtaking decisions is essential for improving traffic flow, safety, and overall efficiency. However, the coexistence of both connected and non-connected vehicles, driven by the high cost of full deployment and the incomplete global adoption of standardized communication standards, has led to the emergence of Mixed Connected and Connectionless Vehicles (MCCV) scenarios. These scenarios complicate lane-changing overtaking decisions, as the unpredictability of connectionless vehicles, particularly in dense traffic, poses significant challenges and increases safety risks. Furthermore, incorporating fairness into decision-making is vital to ensure equitable treatment of all vehicles, which is key to improving road safety and traffic efficiency. To address these challenges, we propose a fairness-aware overtaking decision optimization method for MCCV scenarios, which aims to enhance fairness while improving safety and efficiency in vehicle decision-making. First, a Bayesian network-based fairness assessment method is introduced to quantify fairness under limited data conditions by modeling the probabilistic relationships between vehicle behaviors and fairness outcomes. Second, we develop a left-lane availability detection mechanism based on adaptive maneuver tree search and a fast-lane speed recommendation mechanism grounded in traffic flow analysis. These mechanisms enhance compliance with traffic regulations and improve efficiency by dynamically assessing lane conditions and providing a real-time speed recommendation based on traffic density and flow. Finally, we incorporate a K-Nearest Neighbor (KNN)-enhanced deep reinforcement learning approach, which integrates a parameterized dueling deep recurrent Q-network with KNN-enhanced experience replay. This approach effectively copes with rare but critical driving conditions, such as unpredictable vehicle behaviors or sudden traffic dynamics, improving decision-making reliability in various traffic scenarios. Extensive simulations demonstrate that the proposed method significantly enhances the fairness, efficiency, and safety of lane-changing overtaking decisions in MCCV scenarios, effectively addressing the unpredictability of mixed-vehicle interactions.

The vehicular environment presents unique challenges, including massive data generation, stringent latency requirements for safety-critical applications, bandwidth limitations, and intermittent connectivity, which make centralized learning approaches impractical. Vehicular Federated Learning (VFL) enables distributed model training by leveraging local data from connected vehicles, while preserving data privacy and reducing network overhead. However, the dynamic nature of VFL presents several additional challenges. High vehicle mobility and unstable channels lead to inconsistent client participation, while heterogeneous vehicle capabilities result in unbalanced training workloads and competitive resource allocation. These challenges significantly degrade VFL model performance and prolong training periods. In this paper, we propose a Mobility Resilient Vehicular Federated Learning (MR-VFL) scheme, which comprises two key components: an amplification-based adaptive vehicular FL (AVFL) training scheme and a dual-timescale FL scheduler. Specifically, AVFL adapts local training epochs to vehicle capabilities to improve scheduling flexibility and alleviate the impact of insufficient local epochs on model updates, which enhances training efficiency and reduces communication competition. The dual-timescale FL scheduler includes a macro scheduling strategy that optimizes long-term VFL performance based on the correlation between convergence speed and model accuracy, and a Mamba-based real-time scheduler that enhances training efficiency and reduces decision latency in massive vehicles scenarios. Extensive simulations show that MR-VFL effectively mitigates performance degradation due to complex vehicle mobility and heterogeneity, and improves training efficiency.

Large language models (LLMs) are revolutionizing various fields due to their powerful generation capabilities. However, their immense computational complexity poses significant challenges in resource consumption, inference latency, and data privacy for traditional cloud-centric deployments. Edge artificial intelligence (Edge-AI) offers promising LLMs deployment solutions by leveraging distributed resources at the network edge. However, existing approaches struggle to adapt to dynamic workloads and efficiently utilize heterogeneous resources in Mobile Edge Computing (MEC) environments. This paper proposes aDynamicBatching andAdaptivePartitioning (DyBAP) scheme for LLMs deployment, which utilizes ubiquitous geo-distributed resources via end-edge-cloud collaboration. Firstly, we formulate a collaboration deployment optimization problem to minimize inference latency and resource usage under heterogeneous resource and user requirements for latency and accuracy constraints, which is NP-hard. Secondly, to solve this, we develop a dynamic batch fusion optimization algorithm that optimizes the batch size of inference by utilizing the parallel processing power of computing units to balance the latency and resource usage. A block-aware partition optimization algorithm based on multi-agent reinforcement learning (MARL) is proposed for efficient transformer block allocation, integrating mobility awareness for optimal partitioning across dynamic network environments. Simulation results demonstrate the superiority of DyBAP over other benchmarks, reducing inference latency by 17.94% and saving 11.12% in memory resource consumption compared to the end-edge-cloud collaboration approaches.

Asynchronous Federated Learning (AFL) can mitigate the straggler problem due to unbalanced training time of clients in Synchronous Federated Learning (SFL), thereby reducing the aggregation time and improving the training efficiency. However, AFL introduces training bias since different updating frequencies of heterogeneous clients can cause unequal knowledge contributions to the global model. Meanwhile, if the client data are heterogeneous, the local optimum may be drifted from the global one, which exacerbates the training bias problem. In order to solve the above issues, we propose a Knowledge Alignment framework for AFL with Heterogeneous Data, termed as KAFL-HD. Firstly, considering data heterogeneity, a data quality-aware aggregation method is proposed to estimate client contributions precisely, where both model staleness and data quality are utilized in aggregation weights. Secondly, a knowledge distillation method with staleness is designed to supplement more knowledge from slow clients to the global model. Thirdly, an adaptive learning rate adjustment method is proposed to customize the local learning rate based on the aggregation frequency and weight, which aligns the knowledge contributions of clients in the local training process. Furthermore, we provide theoretical analysis under a non-convex setting to show the convergence speed of KAFL-HD. Finally, comprehensive experiments are conducted, and the results show that KAFL-HD achieves the highest accuracy and fairness performance compared to the state-of-the-art baselines.

Software-Defined Networks (SDN) bring unprecedented agility and programmability to traditional networks by decoupling the control plane and data plane. However, this separation enables adversaries to manipulate data plane forwarding behaviors or modify packet payloads, thereby violating the network security policies set by the control plane and leading to information leakage, network congestion, or even network collapse. In this article, we propose an Enhanced Lightweight Path Validation Scheme (EL-PVS) for the SDN environment. Firstly, we propose a packet forwarding path validation scheme that verifies the paths traversed by packets, alongside a theoretical analysis of this validation process. Then, we extend the scheme with a network flow-level path validation to improve the validation efficiency, and present a storage optimization method to reduce the storage overhead in the validation process. To support large-scale deployment, we design a path partition scheme and present a Greedy-based KeySwitch Node Selection Algorithm (GKSS) to pinpoint optimal switches for path partition, significantly reducing overall data plane storage usage and the total number of paths requiring validation. In addition, we extend our path validation scheme to detect packet payload modification, where a multi-phase packet modification detection approach is designed, and then the detection results are integrated with path validation information to minimize switch-to-controller bandwidth usage. Finally, we present an anomaly switch identification technique to identify abnormal switches when the controller encounters validation failure. The evaluation results verify that EL-PVS enables flow-level path validation and packet modification detection with small validation header, minimizing processing delay and switch storage overhead.

In cloud-edge-end (CEE) collaboration, the resource optimization based on deep reinforcement learning have achieved significant performance improvements in time-slot systems. However, some studies only focus on computing delay and energy consumption in each time slot, ignoring the impact of task backlog queues on system performance. In addition, the delay-oriented optimization tends to offload a large number of tasks to servers, failing to fully utilize the computing resource of mobile devices. To address these issues, we propose the multi-dimensional preference-driven resource optimization framework (MPROF). This study includes several key points: 1) constructing a three-layer heterogeneous architecture that applying the collaboration among edges for CEE; 2) proposing the task backlog estimation mechanism, which mitigates the impact of previous unfinished tasks on the current time slot; 3) proposing the group relative direct-preference policy optimization (GRDPO) that incorporates the preference information for efficient task offloading, and combines it with mathematical programming for the system resource optimization. The simulation experiments are conducted across multiple typical scenarios. The results show that, the proposed framework outperforms existing mainstream methods in task offloading, system delay, task backlog, and energy consumption control, demonstrating certain practical application prospects.

Mobile Edge Computing (MEC) decentralizes the network's computing and storage capabilities from centralized infrastructure to edge nodes located closer to end-users, enabling context-aware service deployment, low-latency service response, and efficient computation for mobile users. However, achieving energy-efficient service deployment while maintaining service delivery quality remains a significant challenge due to the wide geographic distribution of edge nodes, the dynamic variation of service workloads, and the differences between service deployment and delivery cycles. To address these challenges, we first design a feature encoding strategy and a self-attention-based encoder to extract contextual features, which are fused to support adaptive decision timescale regulation driven by service semantics and system load dynamics. Then, we propose a novel Dual-Timescale Energy-Efficient Service Deployment and Delivery (DT-EESD) framework integrated with hierarchical learning. The upper layer leverages an enhanced decision-making mechanism to optimize proactive service deployment and base station switching on a larger timescale, aiming to reduce long-term network costs. The lower layer employs a fine-grained real-time optimization approach to dynamically handle service delivery and resource allocation on a smaller timescale, effectively responding to dynamic service requests. By incorporating an expected reward-based learning mechanism, the framework efficiently handles the temporal coupling between deployment and delivery cycles. Extensive experiments demonstrate that DT-EESD outperforms baseline algorithms, achieving at least a 10.89% reduction in average system cost while also improving model convergence, reducing delay, and enhancing resource utilization.

Generative AI (GenAI) has become a research hotspot for the task of content creation and production, which suffers from the issue of high latency due to cloud transmission. One effective solution is to integrate serverless computing with mobile edge computing (MEC) to build a communication-efficient GenAI system, where serverless functions are executed via containers on edge servers. However, the nonnegligible latency of container deployment and cold starts degrades the quality of GenAI service. This issue becomes even more serious in dynamic MEC with mobile and uncertain users. In this paper, we study the provisioning of latency-sensitive query services in GenAI-enabled serverless MEC through dynamic utility maximization. While GenAI of users deployed in cloud refers to as primary GenAI, we deploy their GenAI replicas based on serverless functions in edge servers to maximize user service satisfaction (i.e., utility function). We first formulate a joint decision problem, i.e.,GenAIReplicaAllocation andPlacement (GRAP) problem, under various resource constraints. For this problem, we propose an approximation solver with a provable approximation ratio. Then, we consider an dynamic GRAP problem with uncertain values of users and stochastic request arrivals, and devise a performance-guaranteed online algorithm for a special case of the problem by assuming only a small subset of edge servers suffers significant utility degradation. Finally, we conduct theoretical analysis and experimentation to validate the effectiveness of the proposed mechanisms. Experimental results demonstrate that the proposed mechanisms consistently outperform baseline methods in both service latency and user satisfaction.

In Vehicular Edge Computing (VEC), load imbalances among edge servers, driven by varying traffic densities and computational demands across geographic areas, can lead to significant delays, decreased efficiency, and potential service disruptions, adversely affecting both user experience and system reliability. This study proposes an innovative adaptive load balancing method that integrates deep reinforcement learning with predictive analytics to optimize resource allocation in VEC. The framework comprises a predictive model called ST-ChebNet, enhanced with Chebyshev polynomials in graph convolutional networks for accurate workload forecasting, and an adaptive model compression strategy utilizing knowledge distillation to dynamically adjust compression ratios based on anticipated workloads. Additionally, the integration of this predictive model with the Soft Actor-Critic (SAC) algorithm, termed GC-SAC, effectively combines graph-based predictive insights with reinforcement learning techniques to tailor resource distribution, minimizing computational delays and enhancing system responsiveness. The simulation results show that the GC-SAC algorithm can significantly reduce the average delay and average energy consumption of vehicular tasks, as well as the workload rate of edge servers.

To enhance the reliability of rotating machinery, cross-domain fault diagnosis becomes vital for detecting faults under unknown operating conditions. However, multisource domain imbalanced data present significant challenges, as divergent label distributions across domains cause complex domain-class shifts and degrade the performance of cross-domain fault diagnosis. Moreover, diagnostic models often struggle to learn features from minority classes due to label imbalance within each domain, which may degrade the performance in diagnosing these minority classes. To address these challenges, we propose a dual-reweighted Siamese contrastive learning network (DRSC) for cross-domain fault diagnosis with multisource domain imbalanced data. In DRSC, we design a Siamese feature extractor based on a wide-kernel convolutional neural network to capture short-term characteristics and leverage the convenience in extracting domain-invariant features. Subsequently, to alleviate domain-class shifts, we design a reweighted contrastive domain-class alignment mechanism that strategically pulls domain-class pairs together while pushing other health conditions away. Finally, to enable the diagnostic model to learn from minority health conditions, a reweighted health condition classifier is developed by assigning higher weights to the minority classes. Evaluation results on two public datasets illustrate DRSC outperforms comparison models in cross-domain fault diagnosis.

Underwater backscatter is an emerging passive communication technology powered by underwater acoustic energy, which has emerged as a promising solution to the underwater energy problem. However, the backscatter mechanism causes the reflected signal strength to undergo periodic ups and downs due to the phase cancellation effect when the receiving end is in motion. Additionally, during the signal retro-reflective process, the reflected array signals deviate from the optimal beam direction when there is movement at the receiving end, introducing nonlinear attenuation into the signal strength. This paper proposes GKCformer signal strength forecasting method for underwater backscatter systems. It is based on the original sequence and Gram angular field image modal design to add period information. It utilizes Transformer to rearrange and improve channel feature information, and Kolmogorov-Arnold Networks to make the fitting better. Experimental results show that the proposed model outperforms the classical model in mean squared error (MSE) and mean absolute error (MAE) metrics, which demonstrates its better performance in prediction accuracy.

Since its inception, blockchain technology has been characterized by its core attributes of immutability, traceability, and decentralization, which are fundamental to ensuring data security. In the contemporary digital landscape, medical data has emerged as a critical asset, and the integration of blockchain into healthcare has facilitated a range of innovative solutions for secure and efficient data sharing. Beyond its role in data security, blockchain’s smart contracts have attracted significant research interest due to their potential to automate processes and enhance efficiency in medical research and healthcare operations. In this context, this survey provides a systematic and in-depth exploration of blockchain applications in healthcare, with a focus on: (1) analyzing the technical foundations of blockchain and its suitability for healthcare applications; (2) synthesizing the eight key domains where blockchain has demonstrated impact in the healthcare sector; and (3) critically examining the challenges that hinder blockchain adoption in healthcare while identifying future research directions. By presenting a comprehensive review of blockchains transformative potential in healthcare, this survey offers valuable insights for researchers and practitioners engaged in this evolving interdisciplinary field.

In modern vehicular networks, the absence of infrastructure support, such as roadside units (RSUs), presents significant challenges for efficient task offloading and allocation. Limited computational capabilities of individual vehicles, combined with task allocation imbalances caused by varying task complexity and vehicle capacities, further complicate the process. Additionally, the formation of vehicular platoons requires accurate future route and destination information to ensure stable collaboration and effective coordination. However, such information can be challenging to obtain due to dynamic and unpredictable road environments, hindering the reliability of platoon formation. To address these challenges, we propose a platoon-based offloading strategy that integrates deep reinforcement learning (DRL) and long short-term memory (LSTM) networks to enhance task allocation efficiency. This approach also leverages the convoy formation algorithm considering future positions (CFA-FPs) to manage platoon constraints effectively. Experimental results demonstrate that our method significantly improves key performance metrics, including total computation cost, latency, and offloading success rate, compared to other task offloading strategies.

With the recent advancement in network technologies, the vehicle based medical networks extend medical services to mobile vehicles, thereby offering flexible and efficient healthcare services for vehicle users in need. The integration of vehicle based medical network and edge computing enables computation intensive medical service tasks to be offloaded on edge servers, to provide fast service response for vehicle users. An efficient task offloading and resource allocation strategy is critical for Vehicle based Medical Edge Computing System (VMECS) to satisfy real-time and reliability requirements while ensuring service performance. To this end, in this paper, we investigate the problem of task computation allocation in VMECS networks. By introducing deep reinforcement learning, we first present a novel VMECS architecture to automatically achieve the optimal task offloading and resource allocation through the multi-agent collaboration, thereby improving service performance. Then, we formulate the problem of task offloading and resource allocation in VMECS networks as an optimization model with the aim of maximizing task success rate by jointly considering communication interferences, resource allocation and delay requirements. To solve it, we further devise a Distributed distributional deterministic policy gradients based Task offloading and Resource allocation (DTR) algorithm. Final simulation results demonstrate that compared with benchmark algorithms, DTR algorithm can obtain higher task success rate, smaller service time, and less task processing time.

In post-disaster emergency medical rescue operations, rapidly establishing an adaptive and flexible edge computing (EC) network, balancing data offloading with energy consumption, and ensuring the stable operation of the network have become urgent priorities. To address these challenges, we proposed an unmanned aerial vehicle (UAV)-assisted microservice mobile edge computing (MEC) architecture. The architecture can be rapidly deployed to provide temporary network coverage and EC services in disaster-stricken areas. A transformer-based resource management (TBRM) approach is utilized to optimize data offloading efficiency and reduce energy consumption, thereby maximizing the service time of the architecture. To enhance the security and reliability of the architecture, four microservices are designed to manage the full UAV lifecycle, and UAV identity authentication is implemented through dual digital signature certificates. Large-scale simulation experiments have demonstrated the effectiveness of the architecture in complex rescue scenarios, providing strong technical support for postdisaster medical rescue efforts.

Unmanned Aerial Vehicles (UAVs) play a crucial role in next-generation mobile communication systems, serving as aerial base stations to provide services when ground base stations fail to meet coverage requirements. However, trajectory planning and power allocation for collaborative UAVs as Aerial Base Stations (UAV-ABSs) face several challenges, including energy limitations, flight time constraints, high optimization complexity due to dynamic environment interactions, and insufficient decision-making information. To address these challenges, this paper proposes a multi-agent reinforcement learning algorithm, namely Communication Actor Centralized Attention Critic Algorithm (CATEN), to jointly optimize the flight trajectory and power allocation strategies of UAV-ABSs. The proposed algorithm aims to maximize the number of users meeting Quality of Service (QoS) requirements while minimizing UAV-ABSs energy consumption. To achieve this, firstly, an information sharing mechanism is designed to improve the collaboration efficiency among UAV-ABSs. It leverages distributed storage, intelligent scheduling of UAV-ABSs interaction experiences, and gating units to enhance information screening and fusion. Secondly, a multihead attention critic network is proposed to capture correlations among UAV-ABSs from different subspaces. This allows the network to prioritize value information, reduce redundancy, and strengthen UAV-ABSs collaboration and decision-making capabilities. Simulation results demonstrate that CATEN achieves better performance in terms of the number of served users and energy consumption compared to existing algorithms, exhibiting good robustness and adaptability in dynamic environments.

The advent of Multi-access Edge Computing (MEC) has empowered Internet of Things (IoT) devices and edge servers to deploy sophisticated Deep Neural Network (DNN) applications, enabling real-time inference. Many concurrent inference requests and intricate DNN models demand efficient multi-DNN inference in MEC networks. However, the resource-limited IoT device/edge server and expanding model size force models to be dynamically deployed, resulting in significant undesired energy consumption. In addition, parallel multi-DNN inference on the same device complicates the inference process due to the resource competition among models, increasing the inference latency. In this paper, we propose a Resource-aware and Dynamic DNN Deployment (R3D) scheme with the collaboration of end-edge-cloud. To mitigate resource competition and waste during multi-DNN parallel inference, we develop a Resource Adaptive Management (RAM) algorithm based on the Roofline model, which dynamically allocates resources by accounting for the impact of device-specific performance bottlenecks on inference latency. Additionally, we design a Deep Reinforcement Learning (DRL)-based online optimization algorithm that dynamically adjusts DNN deployment strategies to achieve fast and energy-efficient inference across heterogeneous devices. Experiment results demonstrate that R3D is applicable in MEC environments and performs well in terms of inference latency, resource utilization, and energy consumption.

Federated learning (FL) has become a promising technology to provide edge Artificial Intelligence (AI) due to its advantages in privacy protection and reduced communication costs. However, FL is still confronted with privacy leakage issues because the sharing local model may expose the training data information. Existing works typically utilize secure aggregation techniques to eliminate privacy leakage, where local model parameters in FL are obfuscated before they are sent to the aggregator. Nevertheless, secure aggregation makes poisoning attacks more convenient given that existing anomaly detection methods mostly require access to plaintext local models. A Poisoning-Defense Secure Aggregation in FL (PDSA-FL) is proposed to enhance the privacy protection of honest clients and defend against poisoning attacks from malicious clients. Specifically, a Secure Aggregation scheme based on Random Parameters Decomposition (SARPD) is designed to protect client privacy during the FL aggregation process and eliminates the impact of dropped clients on the aggregation results. Secondly, a Poisoning Detection method based on Similarity Grouping (PDSG) is proposed to mitigate the impact of poisoning attacks on the global model of FL without leaking client model parameters. The security analysis discusses the effectiveness of the proposed PDSA-FL in terms of privacy protection. Extensive simulation results show that PDSA-FL can effectively defend against poisoning attacks, significantly improve the convergence performance of global models, and reduce the computation time of clients.

With lithium-ion batteries prevalent in safety-critical industries, accurate and reliable estimation of battery state-of-health, termed trustworthy prognosis, has become crucial. However, neglecting model uncertainty during representative correlation learning may lead to inappropriate aggregation of ambiguous segments. Moreover, shifts in dominant spatial channels and sparsity in useful temporal features further hinder the trustworthy estimation. Furthermore, the misalignment between the predicted and actual distributions leads to a confidence biases issue and degrades performance in distinguishing out-of-distribution samples. To address these challenges, we propose a hierarchical stochastic spatial–temporal Transformer (HSSTT). First, HSSTT implements stochastic self-attention utilizing Gumbel–Softmax reparameterization for uncertainty quantification. Then, a hierarchical spatial–temporal Transformer is designed to leverage uncertainty-aware timestep-wise dilated convolution and clustered stochastic self-attention. Finally, we theoretically analyse the confidence bias issue through bias-variance decomposition and develop a principled calibration strategy. Experimental results on four datasets demonstrate the superiority of HSSTT in trustworthy prognosis against seven State-of-the-Art models.

Vehicle trajectory prediction, as a key enabler of the intelligent transportation system, has attracted considerable attention from academia and industry in recent years. However, the variability and dynamism of traffic conditions pose significant challenges to current vehicle trajectory prediction methods, particularly in the form of domain bias. Domain bias occurs when a model trained on one traffic domain, such as one segment of a highway, underperforms when applied to another segment with different traffic patterns. To address this challenge and advance the field, we propose a new transfer learning-based collaborative vehicle trajectory prediction framework called TRACER, designed to provide reliable and accurate traffic predictions with high adaptability for cross-domain highway traffic scenarios. The core of our framework lies in an adaptive interactive extraction module and a trajectory generation module based on Bidirectional Long Short-Term Memory (BiLSTM), further strengthened by a pre-task of intention recognition for vehicle operation types. To improve model robustness, consistency regularization is applied by injecting disturbances into the target data, and a one-dimensional Convolution (Conv1D)-based intention extraction module is integrated into the BiLSTM-based trajectory generation process, leading to notable improvements in prediction accuracy. Our framework is first trained on source domain data, followed by the transfer of a small amount of labeled data from the target domain, and the overall model is further refined using unlabeled data. By effectively mitigating domain bias, TRACER significantly enhances trajectory prediction accuracy while maintaining high adaptability. The results underscore the importance of addressing domain shift challenges in trajectory prediction tasks and demonstrate the potential of domain adaptation techniques to improve the prediction accuracy of vehicle trajectories across different domains in highway scenarios.

Vehicular Edge Computing (VEC) is an emerging computing paradigm that serves as a specific application of Mobile Edge Computing (MEC) in intelligent transportation systems. As a core technology of VEC, task offloading improves computing efficiency and service quality by offloading computing tasks from vehicular devices to edge nodes. However, the high mobility of vehicles, heterogeneity of resources, and real-time requirements present significant challenges for task offloading. To address the issue of reducing overall system latency and increasing the offloading success rate in multi-task offloading, we propose a task offloading scheme combining Deep Reinforcement Learning (DRL) and Kuhn-Munkres (KM) Matching algorithm, named the PDQKM offloading scheme. Firstly, to mitigate the delay caused by frequent Roadside Unit (RSU) handovers, we propose a method to detect whether a vehicle is within the coverage area of the RSU. This method filters out unreasonable offloading decisions, avoiding the overhead associated with frequent RSU handovers. Secondly, the combination of DRL and the KM matching algorithm leverages the strengths of both approaches. DRL provides initial offloading strategies in highly dynamic and high-dimensional decision environment. Although DRL may get stuck in local optima, it can quickly adapt to environmental changes. The KM matching algorithm, a classic solution for perfect task-resource matching, performs global optimization on the initial strategies provided by DRL. This integration overcomes the limitations that a single algorithm might have. Finally, to effectively coordinate and manage the heterogeneous resources of RSU, we utilize an improved KM matching algorithm to update computational resources in real time, enhancing matching efficiency. Experimental results demonstrate that PDQKM outperforms comparable offloading schemes in terms of overall system latency and offloading success rate optimization.

Vehicle-to-Everything (V2X) technology plays a pivotal role in enabling real-time traffic coordination and safety, warning, and decision support. Within V2X, the Basic Safety Message (BSM) serves as the core to transmit critical vehicle status, location, and intention information to provide a foundation for ensuring reliable traffic safety and coordination mechanisms. Data accuracy stands as a key to the effectiveness and reliability of the V2X system, in which the transmission of error data can potentially result in severe traffic accidents. During vehicular operation, sensors may generate error data owing to looseness or external conditions. However, immediate sensor replacement is often impractical or infeasible. Therefore, this paper introduces a collaborative scheme involving vehicles, Road Side Units (RSUs), and Data Center (DC) to jointly enhance the accuracy of vehicle-transmitted BSMs. Our scheme involves analyzing statistical features of vehicle driving information to detect error BSMs. Subsequently, these detected errors are corrected by leveraging historical data from the vehicle and its relative relationship with surrounding vehicles. In addition, we propose a time optimization method to reduce the average processing time of each data by RSUs. The extensive experimental results demonstrate that the proposed scheme can accurately detect error BSMs and effectively correct error BSMs. The entire scheme also meets the requisite computational latency requirements.

Personalized Federated Learning (pFL) can improve the accuracy of local models and provide enhanced edge intelligence without exposing the raw data in Mobile Edge Computing (MEC). However, in the MEC environment with constrained communication resources, transmitting the entire model between the server and the clients in traditional pFL methods imposes substantial communication overhead, which can lead to inaccurate personalization and degraded performance of mobile clients. In response, we propose a Communication-Efficient pFL architecture to enhance the performance of personalized models while minimizing communication overhead in MEC. First, a Knowledge-Aware Parameter Coaching method (KAPC) is presented to produce a more accurate personalized model by utilizing the layer-wise parameters of other clients with adaptive aggregation weights. Then, convergence analysis of the proposed KAPC is developed in both the convex and non-convex settings. Second, a Bidirectional Layer Selection algorithm (BLS) based on self-relationship and generalization error is proposed to select the most informative layers for transmission, which reduces communication costs. Extensive experiments are conducted, and the results demonstrate that the proposed KAPC achieves superior accuracy compared to the state-of-the-art baselines, while the proposed BLS substantially improves resource utilization without sacrificing performance.

Federated Learning (FL) enables efficient and privacy-preserving Edge Intelligence (EI) in Mobile Edge Computing (MEC). However, implementing FL-enabled EI services faces critical challenges, including data and device heterogeneity, limited network resources, uneven distribution of network infrastructure, etc., which may intensify with increasing system scale. These challenges are particularly acute in multi-provider environments where edge servers are suboptimally allocated across federations, leading to degraded convergence and increased training costs. In this paper, we present a novel Multiple Hierarchical Federated Learning (MHFL) architecture for large-scale FL and design an Edge Server Reallocation scheme (ESR-MHFL) to enhance training efficiency by optimally redistributing edge servers among federations based on their contribution to model convergence. We first develop a closed-form analysis model for MHFL to quantify training time, computation, and communication costs. To improve training efficiency, we analyze the impacts of edge server allocation on convergence and formulate server reallocation as a multi-item auction problem with theoretical guarantees. We then propose ESR-MHFL, which leverages Coalition Structure Generation (CSG) and greedy matching methods to simplify the reallocation problem and enhance efficiency. Extensive numerical simulations demonstrate that ESR-MHFL not only improves model accuracy while reducing training cost but also exhibits strong compatibility with existing client selection methods, achieving improved training efficiency. The total economic expenditure combining all components

Personalized Federated Learning (pFL) can effectively exploit the non-IID data from distributed clients by customizing personalized models. Existing pFL methods either simply take the local model as a whole for aggregation or require significant training overhead to induce the inter-client personalized weights, and thus clients cannot efficiently exploit the mutually relevant knowledge from each other. In this paper, we propose a knowledge-aware parameter coaching scheme where each client can swiftly and granularly refer to parameters of other clients to guide the local training, whereby accurate personalized client models can be efficiently produced without contradictory knowledge. Specifically, a novel regularizer is designed to conduct layer-wise parameters coaching via a relation cube, which is constructed based on the knowledge represented by the layered parameters among all clients. Then, we develop an optimization method to update the relation cube and the parameters of each client. It is theoretically demonstrated that the convergence of the proposed method can be guaranteed under both convex and non-convex settings. Extensive experiments are conducted over various datasets, which show that the proposed method can achieve better performance compared with the state-of-the-art baselines in terms of accuracy and convergence speed.

Software-Defined Networks (SDN) revolutionize traditional networks by separating control and data planes for enhanced agility and programmability. This separation, however, also opens up vulnerabilities, allowing adversaries to manipulate data plane forwarding and breach security policies. To counter this, we propose a Lightweight Path Validation Scheme (L-PVS) specifically designed for SDN environments. Our approach uses a simple validation scheme for packet forwarding paths that verifies the paths traversed by packets. Then, we further amplify the scheme with a network flow path validation to boost the validation efficiency. To reduce storage demands on switches during flow path validation, we develop a storage optimization method that aligns switch storage overhead with network flows rather than individual packets. Furthermore, we formulate a path partition scheme and present a Greedy-based KeySwitch Node Selection Algorithm (GKSS) to pinpoint optimal switches for path partition, significantly reducing overall data plane storage usage. Lastly, we design a technique using temporary KeySwitch nodes to identify anomaly switches when the controller encounters path validation failure. Evaluation results verify that L-PVS facilitates path validation with a reduced validation header size while minimizing the impact on processing delay and switch storage overhead.

Empowered by the large amounts of sensor data in the Industrial Internet of Things, data-driven fault diagnosis has a pivotal role in improving equipment reliability in harsh industrial environments. To enhance diagnostic performance under unknown operating conditions, transfer learning-based cross-domain fault diagnosis has been emerging. However, diagnostic models are prone to overfit to the source domain due to the lack of sample diversity when only a single-source domain is available. Moreover, significant domain shifts between the single-source domain and multiple unknown target domains may degrade the generalization performance on the unknown domains. To address these challenges, we propose a multipseudo domains augmented adversarial domain-invariant learning (MDA-AD) for cross-domain fault diagnosis. First, we design a multipseudo domain generator, where interdomain diversity constraints and manifold-semantic consistency constraints are implemented to avoid overfitting on the source domain by generating diverse and representative pseudo samples. Subsequently, to alleviate the domain shift, we design an adversarial domain-aware classifier that extracts domain-invariant features by introducing an adversarial paradigm between a feature extractor and a domain discriminator. Finally, to further enhance the diversity of the pseudo domains, we implement a diversity-consistency constrained domain-invariant training strategy. The experimental results, obtained through comparative studies, hyperparameter influence analysis, and visualization on two bearing data sets, affirm the superior diagnostic performance of MDA-AD in a single-source domain.

Although semantic segmentation methods have made remarkable progress so far, their long inference process limits their use in practical applications. Recently, some two-branch and three-branch real-time segmentation networks have been proposed to improve segmentation accuracy by adding branches to extract spatial or border information. For the design of extracting spatial information branches, preserving high-resolution features or adding segmentation loss to guide spatial branches are commonly used methods to extract spatial information. However, these approaches are not the most efficient. To solve the problem, we design the spatial information extraction branch as an AutoEncoder structure, which allows us to extract the spatial structure and features of the image during the encoding and decoding process of the AutoEncoder. Border, semantic and spatial information are all helpful for segmentation tasks, and efficiently fusing these three kinds of information can obtain better feature representation compared to the fusion of two types of information in the dual-branch network. However, existing three-branch networks have yet to explore this aspect deeply. Therefore, this paper designs a new three-branch network based on this starting point. In addition, we also propose a feature fusion module called the Unified Multi-Feature Fusion module (UMF), which can fuse multiple features efficiently. Our method achieves a state-of-the-art trade-off between inference speed and accuracy on the Cityscapes, CamVid, and NightCity datasets. Specifically, BSSNet-T achieves 78.8% mIoU at 115.8 FPS on the Cityscapes dataset, 79.5% mIoU at 170.8 FPS on the CamVid dataset, and 52.6% mIoU at 172.3 FPS on the NightCity dataset. Code is available at https://github.com/SXQ-STUDY/BSSNet.

The rapid development of the Internet of Vehicles (IoV) has spurred innovations in Intelligent Transportation Systems (ITS), but it also faces increasingly sophisticated cybersecurity threats. Traditional defense mechanisms often fall short in handling emerging and complex attacks due to the lack of flexibility to adapt to the rapidly evolving IoV environment. An emerging solution is to employ Large Language Models (LLMs), such as ChatGPT, to enhance IoV security, which depends on the quality, quantity, and freshness of the threat data used for fine-tuning. In this paper, we introduce a collaborative vehicular threat sharing framework that utilizes vehicular honeypots to gather threat data for fine-tuning LLMs, thereby bolstering IoV security. Local differential privacy is leveraged to safeguard the vehicles’ privacy. Given that vehicles have different privacy preferences that may change over time, it is critical to design an appropriate incentive mechanism to encourage sustainable participation in the dynamic IoV environment. Moreover, since privacy preferences are the private information of the vehicles, an information asymmetry exists between the vehicles and the IDS cloud server. To address this challenge, we propose a dynamic contract-based incentive mechanism that considers the dynamically changing privacy preference during long-term participation. The optimal contract is derived to maximize the expected utility of the IDS cloud server. Extensive simulation results demonstrate the feasibility of our proposed dynamic contract based incentive mechanism and validate the effectiveness of the LLM-based threat classification in handling complex threats.

The prediction of user trajectories is a fundamental component to support urban traffic management and various advanced transportation applications, such as traffic optimization and location-based services. Trajectory data typically contains multiple behavioral patterns and contexts, including different travel purposes, modes of transportation, time intervals, and geographic regions. These complex factors collectively influence the prediction of user trajectories. However, trajectory prediction models face challenges in effectively distinguishing between these various patterns. In this paper, we propose a novel stack Transformer-based multimodal long-term trajectory prediction (SMTTP) scheme for heterogeneous user behavior patterns. First, a learnable trajectory similarity measure method is proposed to estimate the relative distance between multi-attribute variable-length trajectories. Then, to address the instability of trajectory clustering caused by random initialization, a cluster head initialization algorithm based on high confidence nodes is developed to improve clustering stability and reduce convergence time. In addition, a Transformer-based trajectory prediction model with multi-dimensional feature fusion is proposed to achieve accurate and efficient long-term trajectory prediction. Experimental results on the real telecom dataset in Shanghai, China show that the proposed SMTTP scheme can achieve improved performance in trajectory prediction in terms of prediction error, and also has high accuracy and stability in unsupervised trajectory clustering.

Intelligent Transportation Systems (ITS) aim to enhance traffic management by improving connectivity and data sharing among vehicles and road infrastructure. In a Mixed Connected and Connectionless Vehicles (MCCV) scenario consisting of connected vehicles equipped with On-Board Units (OBUs) and non-connected vehicles lacking OBUs, communication disparities create challenges in critical lane-changing overtaking decisions. These discrepancies hinder the adaptation of fully connected scenarios to dynamic interactions among these different types of vehicles. Considering the diversity in decision-making ways and capabilities of non-connected vehicles in MCCV scenarios, ensuring the coordinated execution of safe and efficient lane-changing overtaking maneuvers by multiple connected vehicles is crucial for enhancing traffic efficiency. Therefore, we propose a collaborative strategy to facilitate safer and more efficient lane-changing overtaking maneuvers for connected vehicles in the MCCV scenario. First, we design a multi-criteria priority detection, and a dynamic event-triggered mechanism based on confidence intervals to foster efficient collaboration among connected vehicles, optimizing decision-making and reducing conflicts. Second, to accommodate diverse driving styles of autonomous and human-driven vehicles, we introduce an Improved Dynamic Precise Fuzzy C-Means (IDP-FCM) algorithm to dynamically identify and adapt to different driving styles, thereby improving safety. Finally, tackling the challenge of multiple connected vehicles performing lane-changing overtaking involving hybrid action space, our proposed Multi-agent Contrastive Parameterized Dueling Deep Q-Network (MCPDDQN) algorithm incorporates contrastive learning to improve strategy stability in complex driving scenarios. Experimental results demonstrate the effectiveness of our strategy in improving road safety and traffic efficiency of the MCCV scenario.

As a distributed learning paradigm, federated learning (FL) can be applied in mobile edge computing (MEC) to support real-time artificial intelligence by leveraging edge computation resources while preserving data privacy in the end devices. However, the unpredictable wireless connections between end devices and edge servers in MEC (e.g., frequent handovers and unstable wireless channels) may result in the loss of important model parameters, which slows down the FL training process and degrades the quality of the global model. In this paper, we propose an adaptive collaborative federated learning (ACFL) scheme to accelerate the convergence and improve model reliability by mitigating communication-based parameter loss under a three-layer MEC architecture. First, a dynamic epoch adjustment method is proposed to reduce communication rounds by dynamically adjusting the training epochs in end devices. In addition, to accelerate the FL convergence, we present an edge server collaborative training scheme by leveraging a multi-layer computing architecture, where edge servers utilize their maintained data to collaboratively train models with end devices. Finally, extensive simulations are conducted and show that ACFL can efficiently improve model reliability and accelerate the convergence of the FL process in MEC.

Due to the poisonousness, explosiveness, and diffuseness of some continuous objects (e.g., toxic gas, nuclear radiation, and industrial dust), continuous object tracking has a pivotal role in protecting the safety of the people, especially in hazardous industries. To improve production safety, the Industrial Internet of Things (IIoT) has become a promising technology for continuous object tracking. However, IIoT can hardly satisfy the requirements of both energy efficiency and tracking accuracy due to diffusion characteristics, redundant packets, unnecessary awakened nodes, etc. To address these challenges, we propose a two-stage continuous object predictive tracking scheme based on a state transition model (TCOT-STM). First, the predictive tracking process of TCOT-STM is partitioned into two stages to determine wake-up regions where the future continuous objects are located. Considering the high diffusion speed in the tracking process, stage I tracking is designed by communication range calibration and global wake-up region establishing. To eliminate the redundant boundary nodes in the tracking process, stage II tracking is designed by intercluster gap eliminating, virtual node generating, and local wake-up region establishing. Then, a state transition model (STM) based on finite state machines is designed to awaken nodes selectively. Finally, with the STM and the wake-up regions determined by two-stage tracking, the potential boundary nodes are proactively awakened for predictive tracking. Simulation results demonstrate that the proposed TCOT-STM can reduce energy consumption and communication cost while improving tracking accuracy.

In Mobile Edge Computing (MEC) networks, dynamic service migration can support service continuity and reduce user-perceived delay. However, service migration in MEC networks faces significant challenges due to the uncertainty in future traffic demands, the distributed architecture of MEC networks, high operating costs and the dynamism of network resources. Digital Twins (DT), which achieve the mapping of physical entities to virtual digital models in cyberspace, provide new perspectives for intelligent and efficient service provisioning in MEC networks. In this paper, we propose a traffic prediction-assisted federated deep reinforcement learning scheme to efficiently migrate services and improve the cost efficiency of DT-enabled MEC networks. Specifically, to address the coupled spatio-temporal dependencies of mobile traffic and the imbalance in traffic data, a Multi-order Spatio-temporal information integration-based distributed Traffic Prediction (MSTP) scheme is proposed, which achieves high-accuracy mobile traffic prediction at a low cost. Then, we propose a Federated Cooperative cost-efficient Service Migration (FCSM) algorithm that adaptively adjusts service migration strategies in a distributed manner to respond to future traffic demands. Moreover, a theoretical model is developed to analyze the convergence of FCSM and derive the upper bound of the time-average squared gradient norm. Finally, extensive simulations demonstrate that the proposed schemes achieve excellent traffic prediction performance, enhance users’ Quality of Service (QoS), and significantly reduce the system cost of MEC networks.

To improve the reliability of autonomous vehicles, open-set fault diagnosis is indispensable to jointly detect known and unknown faults, in which unknown faults only appear in the testing set. However, in learning the representations for open-set diagnosis, the extracted representations lack hierarchy to preserve high-level and genuine representations, and the final representations utilized for diagnosing lack distinctiveness to separate unknowns from knowns. In addition, in the stage of testing, the open-set diagnosis models are error-prone when unknowns are similar to knowns. Motivated by these challenges, we propose a Multi-hop Attentive Graph Variational Autoencoder (MAGVA) model for open-set fault diagnosis in this paper. First, a multi-hop attentive graph convolutional network is developed to adaptively extract hierarchical representations and eliminate unknown fault misidentification. Then, to avoid unknown faults occupying the same region as known faults and identify known faults, structural representation constraints are designed by jointly conducting reconstruction with an intra-class constraint and classification with an inter-class constraint. Finally, combining the distinguishable representations learned by MAGVA, a generative distance-based open-set diagnosis algorithm is proposed, in which the procedures of estimating class-conditional distributions are designed, and a relative generative distance is then presented to derive diagnosis results under the class-conditional distributions. Experiments on three commonly used bearing datasets for vehicles demonstrate that the proposed MAGVA consistently outperforms the compared models in open-set, closed-set, and unknown fault diagnosis.

Mobile-edge computing (MEC) sinks computation and storage capacities to network edge, where it is close to users to support delay-sensitive services. However, due to the dynamic and stochastic properties of MEC networks, the deployed services may be frequently migrated among edge servers to follow the mobility of users, which greatly increases the network operational cost. In this article, considering the service migration cost brought by user mobility, we study the joint optimization problem of service deployment and request routing decisions to maximize the long-term network utility of MEC networks. First, we propose a Lyapunov optimization-based online service migration algorithm to decompose the continuous optimization problem into a number of one-slot online optimization problems. Then, to address the NP-hard issue of one-slot optimization, we use a randomized rounding technique to implement service migration and request routing. Furthermore, through a closed-form theoretical analysis, we prove that the proposed algorithm not only greatly meets the local user requests and enables approximate performance guarantees but also adaptively balances the service migration cost and system performance online. Finally, extensive simulations are conducted, which demonstrate that our algorithm can efficiently utilize the storage and computation resources of edge servers, and maximize the long-term network utility while ensuring the stability of service migration cost.

The Internet of Things (IoT) technology has expanded network space by interconnected devices, which has been widely used in various fields, such as environmental monitoring, object tracking, risk warning, etc. Due to insufficient computing capacity, limited battery life, and unreliable communication environment in IoT, unmanned aerial vehicle (UAV)-enabled edge computing has been recently utilized to provide enhanced coverage and efficient computational support in the scenarios with sparse or unreliable ground infrastructure, such as disaster rescue, emergency response, military fields, etc. However, UAV-enabled edge computing faces many challenges, such as low offloading efficiency, high energy consumption, high complexity, etc. In this article, a distributed computation offloading scheme is proposed to provide computational support to large-scale IoT nodes and optimize the energy efficiency of multiple UAVs. First, to provide accurate and efficient computational support, a real-time intelligent positioning algorithm is designed to obtain the precise location information of IoT nodes. Then, a distributed computation offloading and path planning algorithm is presented, which jointly optimizes the computation offloading of large-scale IoT nodes and trajectory planning of multiple UAVs to reduce the energy consumption of UAVs. Furthermore, we develop a closed-form theoretical analysis model to demonstrate that the algorithm enables a performance guarantee related to energy efficiency. Finally, extensive simulations have been conducted and show that the proposed scheme can greatly improve the system utility and energy efficiency.

The unprecedented development of intelligent manufacturing requires to customize and change the network traffic strategies frequently. With the advantages of highagility and programmability, software-defined networking can dynamically manage industrial networks, which makes it a promising networking technology for intelligent manufacturing. However, the software-defined industrial network architecture is vulnerable to network attacks, which may degrade manufacturing productivity, and even cause accidents. In this article, we propose a deep learning-based one-class intrusion detection scheme (DO-IDS) to improve the security of industrial networks. Firstly, DO-IDS periodically extracts the flow statistics of the industrial network traffic to generate network status features. Then, it utilizes a deep learning-based dimension reduction approach to filter redundant features. In addition, a deep learning-based one-class detector is designed to calculate the abnormal scores of the network status features. Finally, we conduct extensive simulations, which demonstrates that DO-IDS can detect abnormal traffic with enhanced accuracy and high efficiency.

This paper investigates the reliable computation offloading in vehicular edge computing (VEC) systems. Compared with the traditional task replication method in which task replicas are typically assigned to multiple service vehicles at the same time, in our work, a task vehicle allocates the computation tasks and communication resources to its neighboring service vehicles through the vehicle-to-vehicle (V2V) links, and avoids the degradation of delay and computation efficiency. Specifically, an optimization problem is formulated to minimize the task completion delay and ensure offloading reliability. Then, an algorithm based on the penalty and the concave-convex procedure (CCCP) method is proposed to effectively solve the formulated optimization problem. The simulation results show that the task completion delay of the proposed algorithm is only 30% of that in the traditional task replication method.

With the explosive growth of mobile devices at the edge of mobile social networks (MSNs), the amount of the content that needs to be transmitted is exploded. Traditional content delivery mechanisms leverage only local information to make routing decisions, which results in both high latency and low delivery rate. Software-defined networking (SDN) is a novel network paradigm, the design philosophy of which could be applied to MSN for improving the content delivery performance. In this article, the centralized control thought of SDN is introduced into MSN to efficiently process social information. The classical routing algorithm of BubbleRap is improved from the perspective of network density, which is the basis of designing the sparse and dense routing mechanisms for MSNs. In addition, flexibly switching between these two routing mechanisms is implemented by a discriminating scheme, achieving efficient yet adaptive routing. The experimental results show that the delivery ratio of sparse routing is up to 83%, and the dense routing could reach up to 93%.

Wireless communication demands increase sharply with the explosive growth of mobile devices. The communications mainly depend on the infrastructure-based networks, e.g., WLANs and cellular networks. However, such wireless connections may be unavailable in crowded areas (e.g., concert and conference hall) or interrupted by infrastructure failures caused by earthquake or tsunami. These promote the evolution of local communication systems over device-to-device communication, such as Bluetooth and Wi-Fi Direct (WFD). However, none of the existing studies construct a full-featured local communication system, and they do not consider how to support the user mobility either. In this article, we implement and evaluate the performance of a WFD-based local communication system. First, we improve the intragroup communication by the native implementation of WFD on the Android platform, and propose an application-layer forwarding solution for the intergroup communication, which can be applied to three or more connected groups. Then, we put forward a self-adaptive handover mechanism taking user mobility and node failures into account. To deal with the uncertainty in the handover decision procedure, a fuzzy-logic-based normalized quantitative decision algorithm (FNQD) with the weights derived from the fuzzy analytic hierarchy process (FAHP) is utilized. Finally, we evaluate the performance of the system through both simulation and experiment analysis. Results show that we can get a maximum throughput of 31.7 Mb/s for the intragroup communication and a maximum goodput of 4.76 Mb/s for the intergroup communication. What is more, mobile devices could perform various types of handover according to their roles and status, which could improve the robustness of the local communication system.

As an emergent Internet-of-Underwater-Things (IoUT) system, the underwater wireless networks (UWNs), especially the autonomous underwater vehicle (AUV)-based UWNs are considered to be future of deep-sea exploration. Instead of underwater exploring or data collection based on an independent AUV, the multi-AUVs cooperative system or the AUV flock-based UWNs perform more efficiently and accurately in some particular underwater exploring tasks. In this article, we focus on improving the scalability or controllability of the AUV flock-based UWNs and utilize the paradigm of software-defined networking (SDN) to improve the flexibility and controllability of the AUV flock-based UWNs. With the proposed SDN-enabled architecture for the AUV flock-based UWNs, the UWNs are divided into three layers, and the data transmission, synchronization, and collection among the AUVs are implemented by the proposed software-defined beacon and control frameworks. By the centralized management feature of SDN, we define the concept of AUV flock and the united control model based on the artificial potential field theory. Then, we propose an exact path planning scheme for the AUV flock, especially when potential underwater obstacles or “no-go” areas are taken into account. We will show how an SDN controller can be a director/leader for the AUV flock-based UWNs to perform an exact underwater path planning mission. The simulation results show that our proposal is efficient in managing the operation of the AUV flock, especially the proposal SDN controller-guided path planning scheme performs more efficiently than the normal-distributed path planning scheme.

Driven by the emerging advanced information and communication technologies, e.g., artificial intelligence, 5G wireless communications, big data analytics, etc., industrial Internet serves as a key enabling technology to realize intelligent manufacturing, and has been attracting considerable attentions from academia and industry. However, the traditional industrial networks can hardly satisfy the quality of service (QoS) requirements for some mission-critical industrial applications (e.g., fault detection, advanced control, remote monitoring, predictive maintenance, etc.) due to network heterogeneity, traffic congestion, dynamic end-to-end latency, reliability issues, and so on. The emerging software-defined networking (SDN) has been considered as a promising architecture to improve the QoS of industrial applications by flexibly decoupling the control and data planes to control the network behaviours centrally. Owing to economy and policy considerations, a realistic solution is to incrementally deploy SDN in industrial networks instead of fully replacing traditional industrial routers with SDN-enabled switches. In this article, we consider a hybrid Industrial network consisting of conventional routers (e.g., running OSPF protocol) and SDN-enabled switches (e.g., running OpenFlow protocol), and propose an intelligent QoS-aware forwarding strategy to improve the QoS of industrial applications, by utilizing a single path minimum cost forwarding scheme and a K-path partition algorithm for multipath forwarding. Simulation results demonstrate that the proposed scheme not only guarantees the QoS requirements of industrial services, but also efficiently utilizes bandwidth resources by balancing traffic load in the SDN/OSPF hybrid industrial Internet.

With the development of deep learning methods such as convolutional neural network (CNN), the accuracy of automated pulmonary nodule detection has been greatly improved. However, the high computational and storage costs of the large-scale network have been a potential concern for the future widespread clinical application. In this paper, an alternative Multi-ringed (MR)-Forest framework, against the resource-consuming neural networks (NN)-based architectures, has been proposed for false positive reduction in pulmonary nodule detection, which consists of three steps. First, a novel multi-ringed scanning method is used to extract the order ring facets (ORFs) from the surface voxels of the volumetric nodule models; Second, Mesh-LBP and mapping deformation are employed to estimate the texture and shape features. By sliding and resampling the multi-ringed ORFs, feature volumes with different lengths are generated. Finally, the outputs of multi-level are cascaded to predict the candidate class. On 1034 scans merging the dataset from the Affiliated Hospital of Liaoning University of Traditional Chinese Medicine (AH-LUTCM) and the LUNA16 Challenge dataset, our framework performs enough competitiveness than state-of-the-art in false positive reduction task (CPM score of 0.865). Experimental results demonstrate that MR-Forest is a successful solution to satisfy both resource-consuming and effectiveness for automated pulmonary nodule detection. The proposed MR-forest is a general architecture for 3D target detection, it can be easily extended in many other medical imaging analysis tasks, where the growth trend of the targeting object is approximated as a spheroidal expansion.

To provide satisfactory Quality of Service (QoS) on the move, efficient mobility management is indispensable to provide mobile users with seamless and ubiquitous wireless connectivity. However, both the conventional centralized mobility architecture and the upcoming distributed mobility management face fundamental challenges such as sub-optimal routing, scalability, and so on. The emerging software-defined networking (SDN) architecture can efficiently manage network operations, and accordingly provides a new direction to address the challenges in mobility management. In this paper, we propose an SDN-based Mobility Management (SDN-MM) scheme to support seamless Intro/Inter domain handover with route optimization. SDN-MM decouples mobility management and packet forwarding functions by installing route optimizing and mobility control logics in an SDN controller, but exempting it from traffic redirecting. In SDN-MM, a comprehensive set of signaling operations are designed in order to provide transparent and efficient mobility support for ongoing sessions in each handover scenario, which prevents packet loss and tunneling overhead, and accordingly provide improved QoS to mobile users. For data communications, an SDN controller in SDN-MM pre-calculates the optimal end-to-end route before a handover, and decides whether to migrate traffic to the route by balancing the performance gain and the signaling overhead, which greatly improves bandwidth resource utilization. Finally, we develop a novel analytical model to evaluate the performance of SDN-MM, including signaling overhead, handover latency, and packet delivery cost. The simulation results have been provided to demonstrate that the proposed SDN-MM can greatly improve handover performance and maintain high resource utilization efficiency as well.

Dynamic Host Configuration Protocol (DHCP) provides dynamic use of IP addresses, but it presents challenges to meet smart terminals with great mobility and transient network access patterns. Existing studies have tried to solve this problem through adjusting DHCP lease, which controls how long a host owns an address. However, few studies clearly express the relations among the lease, address utilization and DHCP overhead. In this paper, we uncover how the leases affect address utilization and DHCP overhead with two methods, based on which, we can set the leases for the smart terminals flexibly and judiciously. First of all, we present an emulation technique to evaluate address utilization and DHCP overhead under different leases. It provides an experimental basis for setting the lease for the whole WLAN. Evaluation results show that if the lease is set to 120 min instead of 60 min by default, it can reduce 41.78% DHCP overhead on average and still reserve at least 9.2% address space for the possibly emerging terminals. Then, we model the relationship between the lease and address utilization, as well as the relationship between the lease and DHCP overhead. According to these models, we propose a load-aware DHCP lease time optimization algorithm, which helps to set different leases for each area of the WLAN based on theoretical analysis. Evaluation results show that compared with the default lease for the whole WLAN, a lease combination of {15, 120, 120} for different areas can reduce 36.85% DHCP overhead on average and guarantee there is always 10% available address space.

With ever-rapid development of information and communication technologies, e.g., smart city, industrial Internet, Internet of Things, etc., enormous amounts of data are explosively generated and delivered to the computing center for further processing, which brings additional burden to both transfer components (e.g., the Internet) and data processing units. To efficiently schedule data transfer especially for delay-sensitive traffic, a scalable network architecture with intelligent traffic engineering (TE) policy is indispensable. In this paper, we propose a TE engine DTE-SDN by utilizing the software defined networking (SDN) technology, aiming to schedule the transfer of delay-sensitive traffic. Particularly, with OpenFlow, DTE-SDN captures an overall view of the network in real-time and can monitor the quality of service (QoS) metrics (e.g., throughput and delay) of each network link. In order to schedule the delay-sensitive transfer, a dynamic-scheduling scheme capable of multipath routing is proposed, which allows DTE-SDN to compute the near-optimal scheduling (e.g., path selection and flow distribution), based on the instantaneous QoS metrics. Especially, in the scheduling scheme, we propose a probabilistic-matching approach that aims to distribute the traffic among multiple end-to-end paths according to the computed flow distribution policy and can be deployed in SDN-enabled switches (e.g., Open vSwitch). The simulation results demonstrate that DTE-SDN is able to measure the throughput and delay with acceptable error range and can dramatically enhance the transfer efficiency than the traditional scheduling algorithms.

In vehicular ad hoc networks (VANETs), multi-hop wireless broadcast has been considered a promising technology to support safety-related applications that have strict quality-of-service (QoS) requirements such as low latency, high reliability, scalability, etc. However, in the urban transportation environment, the efficiency of multi-hop broadcast is critically challenged by complex road structure, severe channel contention, message redundancy, etc. In this paper, we propose an urban multi-hop broadcast protocol (UMBP) to disseminate emergency messages. To lower emergency message transmission delay and reduce message redundancy, UMBP includes a novel forwarding node selection scheme that utilizes iterative partition, mini-slot, and black-burst to quickly select remote neighboring nodes, and a single forwarding node is successfully chosen by the asynchronous contention among them. Then, bidirectional broadcast, multi-directional broadcast, and directional broadcast are designed according to the positions of the emergency message senders. Specifically, at the first hop, bidirectional broadcast or multi-directional broadcast conducts the forwarding node selection scheme in different directions simultaneously, and a single forwarding node is successfully chosen in each direction. Then, directional broadcast is adopted at each hop in the message propagation direction until the emergency message reaches an intersection area where multi-directional broadcast is performed again, which finally enables the emergency message to cover the target area seamlessly. Analysis and simulation results show that the proposed UMBP significantly improves the performance of multi-hop broadcast in terms of one-hop delay, message propagation speed, and message reception rate.

In urban vehicular networks, traveling users can enjoy Internet multimedia services through various mobile devices, such as smart phones and laptops. To maintain seamless and ubiquitous Internet connectivity, an efficient handoff scheme has to be employed when mobile users travel across different access networks. However, in the urban vehicular environment, the high velocity of vehicles and the random mobility of users impose great challenges to the design of an effective handoff scheme. In this paper, we propose an Efficient Proxy Mobile IPv6 (E-PMIPv6)-based handoff scheme that guarantees session continuity for urban mobile users. In the registration process, E-PMIPv6 enables mobile users to obtain seamless Internet connectivity either from fixed roadside units or mobile routers and improves cache utilization at the local mobility anchor by merging the binding cache entries of the mobile users. In the handoff process, E-PMIPv6 comprehensively considers various handoff scenarios in the urban vehicular environment and provides transparent network-based mobility support to individual mobile users or a group of users in the same mobile network without disrupting ongoing sessions. In addition, E-PMIPv6 eliminates packet loss by either packet buffering or packet tunneling to improve handoff performance in each handoff scenario. Finally, a detailed analytical model is developed to study the performance of E-PMIPv6 in terms of handoff latency, signaling overhead, buffering cost, and tunneling cost. Analysis and simulation results demonstrate that the proposed E-PMIPv6 successfully extends the scalability of user mobility and greatly improves handoff efficiency in urban vehicular networks.

In order to achieve cooperative driving in vehicular ad hoc networks (VANET), broadcast transmission is usually used for disseminating safety-related information among vehicles. Nevertheless, broadcast over multihop wireless networks poses many challenges due to link unreliability, hidden terminal, message redundancy, and broadcast storm, etc., which greatly degrade the network performance. In this paper, we propose a cross layer broadcast protocol (CLBP) for multihop emergency message dissemination in inter-vehicle communication systems. We first design a novel composite relaying metric for relaying node selection, by jointly considering the geographical locations, physical layer channel conditions, moving velocities of vehicles. Based on the designed metric, we then propose a distributed relay selection scheme to guarantee that a unique relay is selected to reliably forward the emergency message in the desired propagation direction.We further apply IEEE802.11e EDCA to guarantee QoS performance of safety related services. Finally, simulation results are given to demonstrate that CLBP can not only minimize the broadcast message redundancy, but also quickly and reliably disseminate emergency messages in a VANET.

Broadcast is an effective approach for safety-related information exchange to achieve cooperative driving in vehicular ad hoc network (VANET). However, it suffers from several fundamental challenges such as message redundancy, link unreliability, hidden terminal and broadcast storm, etc., which degrade the efficiency of the network greatly. To address these issues, this paper proposes a position based multi-hop broadcast protocol (PMBP) for emergency message dissemination in inter-vehicle communications. By adopting a cross-layer approach considering both the MAC and Network layers in the proposed scheme, the candidate vehicle for forwarding an emergency message is selected according to its distance from the source vehicle in the message propagation direction. Analysis and simulation results show that PMBP can not only quickly deliver emergency messages, but also reduce broadcast message redundancy significantly.

This paper proposes a multi-channel token ring media access control (MAC) protocol (MCTRP) for inter-vehicle communications (IVC). Through adaptive ring coordination and channel scheduling, vehicles are autonomously organized into multiple rings operating on different service channels. Based on the multi-channel ring structure, emergency messages can be disseminated with a low delay. With the token based data exchange protocol, the network throughput is further improved for non-safety multimedia applications. An analytical model is developed to evaluate the performance of MCTRP in terms of the average full ring delay, emergency message delay, and ring throughput. Extensive simulations with ns-2 are conducted to validate the analytical model and demonstrate the efficiency and effectiveness of the proposed MCTRP.

This paper proposes a multi-channel token ring protocol (MCTRP) for inter-vehicle communications (IVC). With MCTRP, the emergency messages can be quickly disseminated with bounded delay, and the desired quality-of-service (QoS) for multimedia traffic can be provided with limited hardware requirement and signaling cost. In addition, MCTRP ensures that all nodes have an equal opportunity to transmit their data to achieve fairness. Through extensive simulations, it is demonstrated that MCTRP can effectively meet the requirements of IVC.