The Mixture-of-Experts (MoE) architecture has emerged as a key enabler for scaling large language models (LLMs), empowering increased model capacity with minimal computational overhead through gating-based dynamic expert activation. However, due to the memory demands introduced by expert modules, MoE inference on resource-constrained devices is still challenging. Existing methods such as model compression and parameter offloading provide partial alleviation but often lead to reduced accuracy or increased latency. In this paper, we propose CasMoE, a general and efficient cascaded framework for accelerating MoE inference on resource-constrained devices. CasMoE employs a two-stage offline-online approach to facilitate efficient expert prefetching. In the offline stage, a parameterized Expert Activation Predictor (EAP) is introduced to accurately predict the corresponding expert activation from the incoming prompt. In the online stage, a non-parametric Expert Activation Matcher (EAM) supporting fast expert retrieval is then integrated with the EAP to form a cascade planner that operates independently of the MoE architecture, predicting activated experts for all MoE layers in a single pass prior to decoding. A gating mechanism is also incorporated to dynamically adjust the sensitivity of the EAM and EAP, enabling a flexible trade-off between inference efficiency and quality. Extensive experiments on diverse downstream tasks demonstrate CasMoE’s effectiveness in accelerating inference while preserving high accuracy.

With the growing demand for seamless global connectivity, the integration of Non-Terrestrial Networks (NTNs) and terrestrial networks presents new opportunities for largescale Federated Learning (FL). To reduce the communication overhead in NTN-assisted FL systems, NTN-assisted Hierarchical Federated Learning (HFL) frameworks have been proposed, which leverage aerial platforms as intermediate aggregation layers to effectively alleviate the communication burden on remote cloud servers. However, the high orbital altitude and mobility of Low-Earth Orbit (LEO) satellites introduce additional communication latency and frequent changes in network topology, making synchronous aggregation at the edge-cloud layer significantly degrade the training efficiency of HFL. Existing hybrid asynchronous HFL frameworks typically rely on static asynchronous triggering strategies that are not well-suited for the dynamic NTN environments and overlook the model drift caused by data heterogeneity among edge nodes. Furthermore, the data and heterogeneity among clients pose further challenges in HFL training performance. To address these issues, we propose a Dynamic Hybrid Asynchronous Hierarchical Federated Learning (DHA-HFL) framework for NTN scenarios. At the edge-cloud layer, we introduce a dynamic semi-asynchronous aggregation mechanism, consisting of an adaptive asynchronous aggregation trigger mechanism and an asynchronous dual-end aggregation strategy. The former adaptively adjusts the global aggregation timing to accommodate the dynamic NTN environments, while the latter enables efficient cloud-edge coordination to mitigate global model drift induced by data heterogeneity and inherent model staleness. At the client-edge layer, we employ a synchronous aggregation mechanism and propose a heterogeneityaware adaptive local iteration control strategy. By deriving the convergence bound of synchronous FL under non-IID data, we design an iterative algorithm to optimize the local iteration count of devices, minimizing training latency at the client-edge layer and accelerating global model convergence. Extensive experiments demonstrate the superior performance of DHA-HFL in terms of training latency, model accuracy, and convergence speed, providing an efficient distributed learning solution for NTN-assisted IoT scenarios.

The vast geographic coverage and sparse customer distribution in rural areas lead to inefficiency in traditional last-mile delivery. Truck-drone collaborative delivery systems have emerged as a promising solution to these rural logistics challenges. Accordingly, we introduce a multi-truck multi-drone collaborative delivery framework. Within this framework, we propose a novel graph embedding module—the edge-gated graph attention network (EG-GAT)—which incorporates multi-dimensional edge features into the attention mechanism and introduces a learnable gating module for adaptive multi-head fusion. We further propose a graded flexible time window mechanism, which permits limited service advancement or deferral while applying graded incentive-penalty structures. This approach better captures the temporal flexibility inherent in rural customer service requirements. The resulting multi-objective truck-drone routing problem is modeled as a rewardmaximization task and solved using multi-agent proximal policy optimization (MAPPO) under a centralized-training with decentralized-execution framework. Extensive experimental results demonstrate that the proposed method outperforms other approaches. Furthermore, studies assess the individual effects of graded flexible time window settings and objective function weight coefficients on the optimization performance of collaborative delivery. Finally, we evaluate the practical advantages of our proposed model using real-world rural road cases.

With the rapid development of industrial automation, few-shot object detection has emerged as a promising solution for recognizing novel categories using only limited annotated data. However, existing approaches often suffer from high computational complexity and limited adaptability when deployed in resource-constrained industrial environments. To achieve precise detection, efficiency, and security, this paper proposes a collaborative computing framework based on an Edge-Cloud Dual-Prototype Graph Convolutional Network (EC-DP-GCN) for few-shot object detection with hierarchical knowledge embedding. The framework comprises three key components: a device–edge–cloud architecture, a Positive-Negative Prototype (PNP) module, and a Class-Prototype-Sample Hierarchical Graph (CPS-HG) module. Specifically, the PNP module explicitly models intra-class diversity by constructing discriminative positive and negative prototypes from limited support samples, thereby enhancing prototype representativeness. In addition, we further introduce the CPS-HG module, which treats the dual prototypes as class-based prior knowledge and models the relationships among samples through a hierarchical graph structure encompassing class, prototype, and sample levels. This design effectively expands the semantic margins in the embedding space to improve knowledge-guided detection. Extensive experiments on the PASCAL VOC and MS COCO benchmarks demonstrate that EC-DP-GCN significantly outperforms strong baselines and previous state-of-the-art methods, achieving an average improvement of 1.1% in 10-shot detection scenarios.

Autonomous systems in VEC-enabled ICVs face many challenges, such as self-organization, privacy breach risks, vehicle selection, and resource allocation. As a distributed training framework, Federated Meta-Learning (FML) provides a powerful tool for adaptive and efficient processing of vehicular tasks while securing vehicle data privacy in VEC-enabled ICVs. However, the high-speed mobility of vehicles leads to higher latency and communication interruptions. This article investigates the vehicle selection and resource allocation scheme, subject to the constraints on the number and the residence time of vehicles, the maximum transmission energy consumption, and the ratio of bandwidth resource allocation. It is proved to be a challenging mixed-integer nonlinear programming problem, and we formulate it as a Markov decision process (MDP). We proposed an adaptive Sum Tree-Deep Recurrent Q-network algorithm (ST-DRQN) to solve the optimal resource allocation. ST-DRQN employs an enhanced empirical selection rule and a proportional priority sampling method to address the problems of inefficient model training and slow convergence. Finally, we conducted experiments using intelligent cars equipped with Raspberry Pi to show the effectiveness of the proposed methodology. Experimental results demonstrate that ST-DRQN achieves adaptability and credibility among ICVs while reducing latency and energy costs incurred by long-term training of FML.

Leveraging flexible deployment and wide-area coverage, Unmanned Aerial Vehicles (UAVs) assisted edge computing (EC) extends computation capability to the network edge and has become a key paradigm for improving the quality of service in the Internet of Things. However, constrained by onboard resources, UAVs struggle to independently process massive heterogeneous tasks in real time. Existing studies mainly focus on cooperation between UAVs and ground devices or the cloud, while the coupled cooperative relationships among UAVs and the performance scalability in large-scale UAV networks remain insufficiently characterized. To this end, we propose a distributed task assignment framework for a multi-UAV cooperative EC system, where each UAV explicitly accounts for cooperation with other UAVs and, leveraging controllable mobility, assigns a portion of local tasks to other UAVs or base stations to minimize the long-term system-wide total latency. To address the resulting joint trajectory control and task assignment optimization problem, we first formulate it as a Markov decision process. We then propose a Multi-head self-Attention critic-assisted MADDPG (MA$^{2}$DDPG) algorithm to train local online decision models for UAVs. Under the fully cooperative setting, we employ a centralized Critic to exploit global information and reduce parameter redundancy; concurrently, a multi-head self-attention mechanism is incorporated into the Critic to aggregate multi-UAV interaction information, delineate implicit cooperative dependencies, and alleviate the network input dimensionality curse arising from increasing UAV scale. Finally, extensive simulation experiments validate the effectiveness of the proposed method.

With the rapid development of vehicular edge computing (VEC) and artificial intelligence (AI), the emergence of vehicle edge intelligence meets the need for real-time vehicle intelligence applications. But the execution of deep neural networks (DNNs) requires a large amount of data input, which results in a large amount of computing resources required for the execution of DNN tasks. This also brings a certain burden to the deployment of DNN tasks and the resource allocation of edge servers. In addition, due to the high mobility of vehicles in the VEC, the backhaul delay of vehicle edge intelligent task results increases, affecting the vehicle’s quality of experience (QoE). We propose a joint optimization strategy for service migration and resource allocation aimed at minimizing the average task completion delay. This strategy comprehensively considers service migration actions and edge server resource allocation, which is proved to be a mixed integer nonlinear programming (MINLP) problem, and hence we formulate it as an Markov decision process (MDP). To solve this problem, we propose a service migration algorithm based on the self-attention mechanism-based double deep Q-network and deep deterministic policy gradient algorithm (SA-DDQN-DDPG) to solve it to obtain the optimal system service migration strategy. The experimental results show that the proposed SA-DDQN-DDPG algorithm has good performance in reducing latency. The average migration latency is reduced by 40.41%, 20.7%, and 14.50% compared with always, DQN and DDQN, respectively.

Magnetic ResonanceImaging (MRI) is widely used in examining and diagnosing prostate diseases due to its high resolution. However, the diverse morphology of prostate tissue presents a significant challenge for precise gland segmentation. Convolutional Neural Networks have demonstrated effectiveness in segmenting prostate regions. Nevertheless, their limited capability in extracting global long-range semantic features often leads to unstable network segmentation performance. To address these challenges, we propose a Deep Transformer-based Vnet framework (DT-VNet), which consists of a symmetric encoder-decoder architecture that explores global contextual features and retains local feature information. To effectively learn global and local features, We propose the Deep Union Transformer (DU-Trans) as an encoding base module for capturing comprehensive information. Additionally, we introduce a Pool Fusion Attention (PFA) module for decoding, which emphasizes learning context dependencies and interaction relationships. PFA can also facilitate the fusion of deep and shallow features. To our knowledge, this is the first study about deep transformer-based Vnet framework for prostate segmentation. We validate and compare our method on several public datasets against current state-of-the-art methods. The results demonstrate the superior performance of our proposed method in segmenting 3D prostate MRI.

There is a rising concern about healthcare system security, where data loss could bring lots of damages to patients and hospitals. As a promising encryption method for medical images, DNA encoding own characteristics of high speed, parallelism computation, minimal storage, and unbreakable cryptosystems. Inspired by the idea of involving Large Language Models(LLMs) to improve DNA encoding, we propose a medical image encryption method with LLM-enhanced DNA encoding, which consists of LLM enhancing module and content-aware permutation&diffusion module. Regarding medical images generally have plain backgrounds with low-entropy pixels, the first module compresses pixels into highly compact signals with features of probabilistic varying and plausibly deniability, serving as another LLM-based layer of defense against privacy breaches before DNA encoding. The second module not only adds permutation by randomly sampling from a redundant correlation between adjacent pixels to break the internal links between pixels but also performs a DNA-based diffusion process to greatly increase the complexity of cracking. Experiments on ChestXray-14, COVID-CT and fcon-1000 datasets show that the proposed method outperforms all comparative methods in sensitivity, correlation and entropy.

Collaborative content dissemination enables vehicles to directly access content from surrounding nodes through Vehicle-to-Everything (V2X) technologies, such as Vehicle-to-Vehicle (V2V) or Vehicle-to-Infrastructure (V2I) communication. This approach significantly alleviates the downlink traffic burden on cellular network base stations caused by repeated downloading of popular content while addressing security challenges in electric vehicle (EV) charging operations, such as payment fraud and data tampering. A key yet unresolved challenge in content dissemination and EV charging is incentivizing vehicles to participate in collaborative processes voluntarily. Existing incentive mechanisms mainly rely on centralized architectures, which are vulnerable to single-point attacks and trust issues in third-party platforms. To overcome these limitations, we propose the PCD-DB (Popular Content Dissemination using DAG-based Blockchain) scheme, which uses a Directed Acyclic Graph (DAG)-based blockchain to incentivize V2X collaboration for dual applications: improving content dissemination efficiency and ensuring secure EV charging transactions. Our novel framework establishes a decentralized incentive system where vehicles act as content propagators or charging service providers, depending on their service capabilities. We define the propagation and charging capabilities of vehicles and use contract theory to design hierarchical contracts tailored to heterogeneous vehicle roles. Numerical results show that our decentralized incentive mechanism significantly improves the efficiency and profitability of vehicle content dissemination while ensuring the security of electric vehicle charging transactions, outperforming existing benchmark methods.

The inevitable trade-off between perceptual performance and communication bandwidth in collaborative perception poses a significant challenge. To mitigate this constraint, we introduceHAFNet, a hybrid-stage collaborative perception method designed for multi-agent collaborative 3D object detection. This method initially generates dense proposal, serving as the Region of Interest (RoI). Next, all proposals are aggregated through the proposed Agent-Foreground List via RoI association. Moreover, secondary sampling is performed according to those foreground regions. Furthermore, we achieve superior feature extraction through geometric and offset encoding. Concurrently, the setting of proxy points effectively reduces the size of the collective perception messages. In the end, those features are fused and interacted to get the detection. Extensive experiments on existing DAIR-V2X and V2V4Real datasets illustrate thatHAFNetalmost keeps a minimum communication bandwidth, while it surpasses existing state-of-the-art methods in 3D object detection tasks.

With the rapid development of intelligent transportation systems, multimodal applications such as autonomous driving and real-time video analytics are increasingly common. Cloud-edge-end computing has emerged as a promising solution to support these latency-sensitive tasks through distributed computing and edge content delivery. However, in urban hotspot areas, limited resources and frequent backhaul transmissions degrade network performance. Optimizing content caching to reduce task execution remains a key challenge. To address this, Unmanned Aerial Vehicles (UAVs) are introduced into vehicular networks due to their low cost and high mobility, serving as aerial base stations to assist ground infrastructure. We propose an cloud-edge-end collaborative caching framework, deploying algorithms on UAVs with computing and storage capabilities, working with Roadside Units (RSUs) and idle vehicles to alleviate resource constraints in hotspots. Within this framework, we apply the Multi-Objective Particle Swarm Optimization (MOPSO) algorithm for UAV deployment optimization. Then, we propose a content request prediction model using Bidirectional Gated Recurrent Unit (Bi-GRU) and attention mechanisms. Finally, a content caching algorithm based on Soft Asynchronous Advantage Actor-Critic with Action Mask Module (SA3C-AM) is introduced to minimize latency. Experimental results show that compared to baseline methods, our approach improves cache hit rate by 15.4%, reduces content fetches by 34.08%, and lowers average request latency by 17.6%.

In recent years, deep neural networks (DNNs) have been widely used in Vehicular Edge Computing (VEC), becoming the core technology for most intelligent applications. However, these DNN inference tasks are usually computation-intensive and latency-sensitive. In urban autonomous driving scenarios, when a large number of vehicles offload tasks to roadside units (RSUs), they face the problem of computational overload of edge servers and inference delay beyond tolerable limits. To address these challenges, we propose an edge-vehicle collaborative inference acceleration mechanism, namely Model partitioning and Early-exit point selection joint Optimization for Collaborative Inference (MEOCI). Specifically, we dynamically select the optimal model partitioning points with the constraint of RSU computing resources and vehicle computing capabilities; and according to the accuracy threshold set to choose the appropriate early exit point. The goal is to minimize the average inference delay under the inference accuracy constraint. Therefore, we propose the Adaptive Dual-Pool Dueling Double Deep Q-Network (ADP-D3QN) algorithm, which enhances the exploration strategy and experience replay mechanism of D3QN to implement the proposed optimization mechanism MEOCI. We conduct comprehensive performance evaluations using four DNN models: AlexNet, VGG16, ResNet50, YOLOv10n. Experimental results show the proposed ADP-D3QN algorithm reduces average inference delay by 15.8% for AlexNet and 8.7% for VGG16 compared to Edgent algorithm.

In urban temporary congestion or hotspot scenarios, fixed roadside units (RSUs) often fail to provide reliable and efficient communication services due to their static deployment. Unmanned Aerial Vehicles (UAVs) can be rapidly deployed to provide flexible and on-demand communication and sensing support, effectively complementing ground infrastructure. However, UAVs are faced with challenges such as limited coverage, high deployment complexity, and unbalanced communication-sensing performance. These challenges give rise to increased energy consumption and reduced communication efficiency. To address these issues, we propose a UAV energy-efficient deployment method based on Integrated Sensing and Communication (ISAC), which balances performance and energy consumption. Specifically, under UAV energy constraints, we jointly optimize UAV deployment positions and beamforming to maximize communication capacity. We decompose the problem into two subproblems: UAV deployment and beamforming strategy optimization. During the iteration process, the subproblems are respectively solved by using the sparrow search algorithm based on refraction-based learning and successive convex approximation-based iterative algorithm and the first order Taylor expansion method. Simulation results show that the proposed method outperforms the benchmark schemes, achieving approximately a 10.51% improvement in average coverage rate and a 19.83% reduction in UAV energy consumption, while maintaining an effective trade-off between communication coverage and energy efficiency.

Incremental few-shot semantic segmentation (IFSS) expands segmentation capacity of the trained model to segment new-class images with few samples. However, semantic meanings may shift from background to object class or vice versa during incremental learning. Moreover, new-class samples often lack representative attribute features when the new class greatly differs from the pre-learned old class. In this paper, we propose a causal framework to discuss the cause of semantic shift and incompleteness in IFSS, and we deconfound the revealed causal effects from two aspects. First, we propose a Causal Intervention Module (CIM) to resist semantic shift. CIM progressively and adaptively updates prototypes of old class, and removes the confounder in an intervention manner. Second, a Prototype Refinement Module (PRM) is proposed to complete the missing semantics. In PRM, knowledge gained from the episode learning scheme assists in fusing features of new-class and old-class prototypes. Experiments on both PASCAL-VOC 2012 and ADE20k benchmarks demonstrate the outstanding performance of our method.

With growing interest in secure storage, decentralized storage systems (DSSs) have attracted attention due to their strong data integrity but remain limited by restricted capacity, low efficiency, and weak security guarantees. These shortcomings make existing DSSs unsuitable as practical file storage platforms, particularly in private cloud scenarios where performance bottlenecks are easily exposed. To address this, we present Haina Storage (HNS), a decentralized storage system redesigned for scalability, efficiency, and security in private clouds. HNS introduces a bi-directional circular linked chain structure in which each file forms an independent chain, eliminating inter-file dependencies and enabling parallel retrieval. We further propose a lightweight consensus mechanism, Proof of Resources, that accounts for both storage capacity and network conditions, ensuring fair and timely data placement. Security is strengthened through dynamic access control, confidential block distribution, and decentralized key protection without trusted third parties. Extensive experiments on both public cloud servers and local clusters demonstrate that HNS achieves performance comparable to IPFS while offering significantly stronger security, theoretical scalability advantages, and an efficient and fair resource-based consensus mechanism. All prototype code11https://github.com/Zijian-Zhou/Haina_Storage and experimental modules22https://github.com/Zijian-Zhou/Haina_Storage_Exp are open-sourced, and a demonstration video33https://youtu.be/2b8JMqvZV60 is available online.

ABSTRACT Recent 3D point cloud classification has predominantly focused on local spatial attention, neglecting distant contextual relationships due to the inherent sparsity of LiDAR‐generated data over longer distances. Existing 3D object detection methods prioritize local features, hindering the extraction of semantic information. Despite attempts with transformers, methods often reduce computations through local spatial attention, neglecting content class and scarcely establishing connections among distant global points. Our proposed point class‐adaptive transformer (PCaT) addresses these limitations by establishing long‐range feature dependencies while significantly reducing computations. PCaT includes three key modules: the class‐adaptive transformer (CaT), which utilizes local self‐attention and global self‐attention based on class similarity to facilitate an efficient trade‐off between capturing extended‐global dependencies and managing computational challenges; nested binary clustering (NbC), which dynamically partitions queries into multiple clusters based on content features in each Transformer block; and the AfA, which aggregates high‐dimensional features using max‐pooling alongside a residual MLP component and low‐dimensional features using average pooling and a CaT block. Additionally, PCaT incorporates point cloud segmentation via local–global feature aggregation (PcSeg) to facilitate effective point cloud segmentation. Extensive experimentation on the ModelNet40, ScanObjectNN, and S3DIS datasets demonstrates the superior performance and reasonable stability of PCaT compared with existing methods. PCaT achieves 94.2% overall accuracy (OA) and mIoU scores of 89.2% and 86.2% for the ScanObjectNN and S3DIS datasets, respectively.

Effective power line inspection is crucial for power grid maintenance and management. To address the issue of signal loss or signal weakness in remote suburbs and deep mountains, we presented a UAV patrolling system with hybrid communication modules and edge detection capabilities. This system first features the communication ability in different scenarios including the extreme signal loss case by incorporating the BeiDou Navigation Satellite System, short-message communication, the Internet of Things, and edge computing. Next, to accurately and timely detect the power line faults, such as exposed wire, thatch covering, and lead stem falling, under various conditions, an edge detection model employing YOLOv8 is proposed without the help of cloud centers and public communication networks. Finally, experiments are designed on our built UAV patrolling test bed with different use cases. Numerical results show that our proposed scheme could efficiently inspect the power line, especially in signal loss or weak areas, and have a high precision and low latency for line fault detection, compared to the YOLOv8 baseline algorithm.

Networks in the 5G and beyond era can use software-defined networks (SDN) to achieve network slicing (NS), so as to meet the extremely diverse service requirements of diverse applications in the Internet of Vehicles (IoV). However, the flow fluctuations in the highly dynamic IoV make it difficult to provide reliable, flexible, and scalable services for the IoV by the SDN control plane. Careful SDN controller placement can be a feasible solution to achieve its robustness and flexibility to deal with the changes in network status. Thus, this paper studies a dynamic controller placement problem to improve the performance of IoV services. To be specific, a hierarchical SDN control plane for the IoV is considered with the SDN controllers placed at the edge of networks. Under this architecture, we model the dynamic controller placement by Markov Decision Process (MDP). To efficiently solve the formulated NP-hard problems, we develop an algorithm based on Deep Deterministic Policy Gradient (DDPG) because of its advantages in solving the problem with multi-dimensional action and large solution space. Further, we incorporate a random process into the action selection strategy of DDPG to prevent it from getting trapped in local optimum. Simulation results show that the proposed DDPG-based controller placement approach can adapt to a highly dynamic IoV environment with outstanding performance.

Recently, fixed-wing unmanned aerial vehicles (UAVs) are able to extend the communications mission time and ease of deployment due to their powerful onboard capabilities and flexibility, and reflective intelligent surfaces (RISs) are capable of reflecting links to avoid obstacles and thus improve channel gain. Therefore, RIS-assisted fixed-wing UAVs are widely used in wireless communications. Nevertheless, fixed-wing UAVs have limited energy, so improving energy efficiency is critical. This article focuses on energy efficiency optimization problems under RIS-assisted fixed-wing UAV communications. In communications coverage systems, the flight trajectory of fixed-wing UAVs and service scheduling to ground nodes (GNs) significantly impact energy efficiency. Existing work often adopts circular trajectory and traditional deep reinforcement learning (DRL) algorithms for optimizing trajectory and service scheduling. However, circular trajectory can not adapted to the GN distribution well. In addition, the traditional DRL algorithm has two drawbacks: 1) the efficiency of exploring the empirical process is low and 2) the accuracy of handling the hybrid action space needs to be higher. Thus, we propose the midpoint iteration convex hull (MICH) algorithm based on the Graham scan to design trajectories that can be adapted to the distribution of the GNs. In addition, we propose the action screening virtual and real experience (AS-VRE) mechanism and the N-steps hybrid deep Q and policy network (NsHQPN) algorithm to address the low-exploration efficiency and the low-fetch accuracy in handling the hybrid action space. Experiments show that our proposed MICH algorithm, AS-VRE mechanism, and NsHQPN algorithm can effectively improve the system energy efficiency and outperform other baseline schemes.

Reconfigurable-intelligent-surface (RIS)-assisted autonomous aerial vehicles (AAVs) communications technology improves energy efficiency by reflecting signals. This article utilizes RIS and deep reinforcement learning (DRL) to optimize the scheduling of ground terminals (GTs), AAV trajectories, resource allocation, and time slot lengths to maximize system energy efficiency. Three flaws of the existing DRL algorithm are also addressed to seek higher energy efficiency further. First, DRL faces exploration challenges due to the complexity of the solution space, resulting in low rewards. We propose the ant colony DRL (ACDRL) algorithm, which optimizes the scheduling order of the GTs using the ant colony optimization (ACO) algorithm and feeds the results back to the DRL to optimize the subsequent decision making, thus reducing the exploration overhead. Second, to reduce the degree of local optimization when dealing with hybrid action space planning, we propose a hybrid discrete-continuous DRL (HDCDRL) algorithm to improve action accuracy. Finally, to better generalize the model to similar tasks, we propose the transfer-DRL (T-DRL) model to reduce the training time when the task changes. Experimental results show that our proposed solution outperforms the benchmark solution.

Mobile Edge Computing (MEC) can accelerate computation-intensive applications and emerge as a promising technology for enabling Internet of Things (IoT). MEC improves the processing performance of tasks by assigning them to the edge nodes. However, with massive terminals contending for computation and communication resources simultaneously, how to develop a flexible computational offloading mechanism becomes the fundamental issue of MEC-enabled IoT systems. This paper aims to develop an effective computational offloading decision scheme by jointly considering the computational resource and diverse user demands with two goals, i.e., minimizing both the latency and the energy consumption. Specifically, we develop a two-stage computational offloading mechanism, where the computational resources and offloading decisions can be allocated and coordinated with the variation of computation requirements. To achieve the two goals, this work introduces an edge node recommendation model within the cloud-edge-end architecture to reduce the offloading optimization search space. Furthermore, we propose a new computational offloading (CROCA) algorithm based on Chemical Reaction Optimization (CRO) for optimizing offloading utility, which thoroughly considers the competition between mobile device requests and computational resources. Extensive evaluation results demonstrate that the proposed CROCA scheme can effectively improve the computational offloading performance.

A critical challenge for predicting urban traffic flows is to simultaneously process time series and spatial features from heterogeneous traffic data collected by diverse Internet of Things (IoT) devices. Despite the advent of Transformer-based models with an advanced network structure and excellent prediction performance, standard Transformer models are still struggling to combine both spatial information and temporal relations of traffic flows. To address these challenges, we design a novel Transformer network, namely temporal-spatial traffic-flow Transformer (TST-Trans), for traffic flow prediction with high accuracy. In particular, we use learnable position encoders to replace traditional fixed position encoders. Meanwhile, we introduce a spatiotemporal embedding method that integrates temporal relationships and spatial information with external inputs, thereby capturing the spatiotemporal dependencies of traffic flows. Experiments with the real-world datasets demonstrate that our proposed TST-Trans achieves better prediction accuracy than state-of-the-art methods while requiring fewer parameters. The research results increased by more than 10% compared with Transformer. Compared to spatiotemporal deep hybrid neural network, there is a 2% to 10% improvement in performance on different datasets.

This paper addresses the challenge of fault analysis in the Vehicular Energy Network (VEN) caused by small fault samples due to transient faults and complex disturbances. The proposed generation and diagnosis networks (GDNs) are developed without necessitating prior knowledge or manual intervention. The approach starts with an encoding and diagnosis network that converts multi-dimensional signals into images through supervised learning. A sample enhancement network, improved with module transfer and a relaxation objective function, is then proposed to increase the reliability of convergence and diversity of features for small fault samples. Additionally, a joint iterative training strategy between these two networks improves diagnostic accuracy and generalization through feature feedback. Performance validation on a semi-physical simulation platform demonstrates that the proposed GDNs achieve a 20% improvement in diagnostic accuracy with small datasets (200 samples) and maintain superior performance as sample volume grows. Thus, the proposed approach offers a potent solution for fault diagnosis in VENs with scarce samples, enhancing the analysis of complex systems. Note to Practitioners—This paper delves into fault diagnosis in the vehicular energy network (VEN) using small samples, employing a data-driven and deep learning model. The proposed method is versatile, suitable for analyzing complex systems with multiple and heterogeneous signals. An end-to-end model, named generation and diagnosis networks (GDNs), is introduced for generating small samples and conducting fault diagnosis without requiring prior knowledge or manual input. This method encodes multiple signals into signal images, which are then processed by a specially designed sample enhancement model, improved through the relaxation objective function and module transfer method. The enhanced samples are utilized for accurate analysis within the encoding and diagnosis networks’ diagnostic unit. The paper also provides a comparison of diagnosis results for reference. This approach enables researchers and engineers to efficiently augment and analyze small samples in practical applications, offering a practical framework for junior and inexperienced analysts. Preliminary experiments conducted using the RT-Lab semi-physical simulation platform suggest the method’s feasibility and effectiveness. Future research will explore the optimization of the model’s topology and parameters for lightweight design.

Cloud gaming represents a major part of contemporary gaming. To boost the Quality-of-Experience (QoE) of cloud gaming, the integration of Dynamic Adaptive Video Encoding (DAVE) with Multi-access Edge Computing (MEC) has become the natural candidate owing to its flexibility and reliable transmission support for real-time interactions. However, as multiple gamers compete for limited resources to achieve personalized QoE, such as ultra-high video quality and ultra-low latency, how to support efficient edge resource optimization is a fundamental and important problem. Furthermore, determining the optimal game video encoding configuration in real-time poses significant challenges, especially when lacking the information on future video and edge network resources. To address these key issues, we jointly optimize the video encoding as well as computing and communication resource allocation by active mutual adaptation of video coding configurations and physical resources in a Software Defined Networking (SDN)-assisted edge network. This eliminates the performance bottleneck caused by decoupling optimization of coding parameter configuration and physical resource allocation. The SDN-assisted edge network architecture supports efficient on-demand resource management, provides global network information, and meets the stringent time-varying game requests. Due to the significant time scale difference between video chunk and physical resource block, we propose a novel Asynchronous Decision-Making Multi Agent Proximal Policy Optimization algorithm (AD-MAPPO), which can address the credit assignment problem with a single agent. It can also adapt to the highly dynamic cloud gaming environment without prior knowledge and a deterministic environmental model. Extensive experimentation based on real cloud gaming datasets convincingly demonstrates that our approach can significantly enhance the overall QoE of gamers.

Unmanned Aerial Vehicle (UAV)-assisted Vehicular Edge Computing (VEC) has emerged as a novel communication paradigm for compute-intensive and latency-sensitive applications in Intelligent Connected Vehicles (ICVs). In urban hotspots with traffic congestion, UAVs can alleviate the overload on edge servers, enhancing the ICVs’ quality of service (QoS) and safety. However, due to the high mobility of vehicles, effective solutions that can support low-latency and higher QoS for ICVs still lacked. Based on this, we first investigated the UAV deployment problem of maximizing vehicle coverage and system energy efficiency, and we proposed an adaptive population differential evolution (APDE) algorithm to address it. Then, we investigate the service caching and computation offloading problem of minimizing the system energy consumption and task response delay. DRL has effectively solved sequential decision-making in dynamic ICV environments, and we proposed a DRL-based Bias Correction-A3C-Gradient Sharing (BC-A3C-GS) algorithm to address it, which improves the traditional A3C algorithm in bias correction and global model update. Finally, experiment results show that the proposed UAV deployment algorithm can bring about 11.58% vehicle coverage rate improvement, and BC-A3C-GS algorithm can decrease about 30.08% task response delay and 17.4% energy consumption, and increase about 10.73% service cache hit rate.

As a distributed embedded system, vehicular edge computing (VEC) completes various complex Deep neural network (DNN) tasks through network collaboration and communication. However,due to the limited computing power of vehicle processors, vehicles cannot handle increasingly complex DNN tasks. To accurately estimate the execution latency of each layer across different DNN models on heterogeneous devices, we proposed the Extreme Gradient Boosting Tree (XGBoost) algorithm to predict DNN task inference latency. Furthermore, we proposed partitioning and offloading algorithms for both chained DNN tasks and Directed Acyclic Graph (DAG)-type DNN tasks, addressing their unique computational characteristics. For chained DNN tasks, we employ a linear search to determine optimal partitioning points based on predictions from the DNN latency prediction model. For the partitioning and offloading of DAG-type DNN tasks, we construct it as a minimum cut problem under the network flow graph and propose a DNN task partitioning and offloading algorithm based on the highest label pre-stream push (HLPP) algorithm to effectively reduce the cost of task partitioning and offloading. Finally, we used an experimental vehicle equipped with Raspberry and a RSU equipped with Jetson Nano to verify the results. The experiment shows that the DNN latency prediction model based on the XGBoost we proposed can effectively improve the latency prediction accuracy of DNN layer-by-layer execution. At the same time, the division and offloading algorithms for different types of DNN inference tasks can achieve higher task completion rate, lower latency, and lower energy consumption.

The development of novel applications causes increased demands on the computational capabilities of Vehicular Edge Computing (VEC). Current works have introduced Unmanned Aerial Vehicles (UAVs) into VEC to solve the resource-constrained problem. However, given the limited storage of UAVs, the key question is to design the offloading strategy and determine which service programs should be cached. In this article, we propose a three-stage game model that aims at providing a precise analysis of the interaction among the Base Station (BS), the UAV, and the User Vehicle (UV). In stage I, the BS is responsible for determining the cache strategy of the UAV and communicating the price strategy to the UVs. In stage II, the UAV communicates the price strategy to the UVs. In stage III, each UV determines its offloading decision based on the price strategy, to minimize the task execution delay and cost. Compared with current approaches, we cache the frequently requested services in the UAV to satisfy the real-time requirements and use game theory to solve the decision-making, which achieves the effect of reducing the delay and cost. The experiment results are performed to assess the convergence and effectiveness of the proposed algorithm.

With the development of medical technology, ultrasonography has become an important diagnostic method in doctors' clinical work. However, compared with the static medical image processing work such as CT, MRI, etc., which has more research bases, ultrasonography is a dynamic medical image similar to video, which is captured and generated by a real-time moving probe, so how to deal with the video data in the medical field and cross modal extraction of the textual semantics in the medical video is a difficult problem that needs to be researched. For this reason, this paper proposes a pre-training model of multimodal distillation and fusion coding for processing the semantic relationship between ultrasound dynamic Images and text. Firstly, by designing the fusion encoder, the visual geometric features of tissues and organs in ultrasound dynamic images, the overall visual appearance descriptive features and the named entity linguistic features are fused to form a unified visual-linguistic feature, so that the model obtains richer visual, linguistic cues aggregation and alignment ability. Then, the pre-training model is augmented by multimodal knowledge distillation to improve the learning ability of the model. The final experimental results on multiple datasets show that the multimodal distillation pre-training model generally improves the fusion ability of various types of features in ultrasound dynamic images, and realizes the automated and accurate annotation of ultrasound dynamic images.

3D object detection by autonomous vehicles is integral to intelligent transportation. Existing systems often compromise essential foreground point features and local spatial interactions through random down-sampling, focusing primarily on local feature extraction. However, this neglects interactions among distant yet significant points, limiting semantic information and detection performance due to inherent point cloud data sparsity. Addressing this, our proposed Geometric Feature Aggregation and Self-Attention in a Multi-Head Transformer (GFA-SMT) architecture leverages Graph Convolutional Networks and multi-channel transformers to enhance weak semantic information of distant sparse objects. GFA-SMT comprises three modules: Distance Suppression for Local Receptive Fields (DsLRF), Geometric Feature Aggregator with Multi-head Self Attention (GFaSA), and Predicted Key-point Weighting and Refinement (PKwR). DsLRF preserves foreground features, GFaSA encodes similar features and aggregates edge features, while PKwR focuses on key-points for enhancing geometric knowledge of distant and sparse objects. Extensive experiments on KITTI, DIARV2X-I and NuScenes datasets show significant enhancements in widely used techniques, resulting in notable increases in average precision (AP) for 3D object detection: 4.08%, 5.56%, and 4.62%, respectively, on the KITTI test dataset. GFA-SMT enhances point cloud detection accuracy, particularly at medium and long distances, with minimal impact on run-time performance and model parameters.

Federated learning (FL) provides an applicable solution for computation offloading in Unmanned Aerial Vehicle(UAV)-assisted Vehicular Edge Computing (VEC) by preserving privacy. However, the heterogeneity of clients brings challenges to the generalization of models. Therefore, we propose a federated meta-learning (FML) framework to solve computation offloading for UAV-assisted VEC. In this paper, we are concerned with computation offloading of temporary hotspot regions due to traffic congestion. Firstly, we construct a computation offloading problem with energy-delay tradeoffs and convert the problem to a Markov Decision Process (MDP). Then, we use FML to train personalized models for different vehicles while enhancing the generalization, we propose a Graph neural network-based FL Probabilistic Embedding for Actor-critic RL (GFL-PEARL) algorithm. We model the context as a Directed Acyclic Graph (DAG) and use GNN to reconstruct the inference network of the PEARL algorithm to extract the correlation between contexts fully. We dynamically adjust the task priority during the FML training process to improve the sampling efficiency. Finally, we verify the performance of the algorithm through simulation and physical experiments. Experimental results show that our algorithm can reduce average cost and task overtime rate by 31% and 56% respectively compared with the benchmarks.

As a promising vehicular communication technology, Cellular Vehicle-to-Everything (C-V2X) is expected to ensure the safety and convenience of Intelligent Transportation Systems (ITS) by providing global road information. However, it is difficult to obtain global road information in practical scenarios since there will still be many vehicles on the road without onboard units (OBUs) in the near future. Specifically, although C-V2X vehicles have sensors that can perceive their surroundings and broadcast their perceived information to the C-V2X system, their line-of-sight (LoS) is limited and obscured by the environment, such as other vehicles and terrain. Besides, vehicles without OBUs cannot share their perceived information. These two problems cause extensive areas with unperceived information in the C-V2X system, and whether vehicles are in these areas is unknown. Thus, extending the perceivable range of the limited scenario for C-V2X applications that require global road information is necessary. To this end, this paper pioneers investigating the scenario inpainting task problem in C-V2X. To solve this challenging problem, we propose an effiCient trAfficScenario inpainTing (CAST) solution consisting of a generative architecture and knowledge distillation, simultaneously considering the inpainting precision and computation efficiency. Extensive experiments have been conducted to demonstrate the effectiveness of CAST in terms of Precise Inpaint Rate (PIR), Rough Inpaint Rate (RIR), Lane-Level Inpaint Rate (LLIR), and Inpaint Confidence Error (ICE), paving the way for novel solutions for the inpainting problem in more complex road scenarios.

As the demand for global computing coverage continues to surge, satellite edge computing emerges as a pivotal technology for the next generation of networks. Unlike ground-based edge computing, Low Earth Orbit (LEO) satellites face distinctive challenges, including high-speed mobility and resource limitations, etc. Therefore, effectively utilizing LEO satellites for global coverage services is crucial but challenging due to their dynamic coverage areas and diverse task requirements. To address these challenges, we introduce a novel dual-cloud edge collaborative task offloading architecture in the satellite network-assisted edge computing environment, namely,Satellite-GroundTaskOffloading (SGTO). The architecture employs a Geostationary Earth Orbit (GEO) satellite and a ground cloud computing center as satellite cloud and ground cloud, respectively, and LEO satellites as edge nodes. We formally define the task offloading problem in theSGTOwith the aim of minimizing the average latency and average energy consumption. We then propose an adaptive approach namedSGTO-Afrom the perspective of satellites to adaptively solve the problem leveraging deep reinforcement learning. Specifically, we transform the task offloading problem into a Markov decision process and adopt the generalized proximal policy optimization (GePPO) algorithm to solve the problem. Finally, experimental results demonstrate thatSGTOarchitecture andSGTO-Aoutperform the representative approaches in terms of average latency, average energy consumption and running time.

As autonomous driving technology advances, the intelligence levels of vehicles continue to increase. However, meeting the demands of autonomous driving in various scenarios requires improved wireless communication and vehicle perception capabilities. Integrated sensing and vehicular edge computing (VEC) technology can provide collaborative perception and computing resources for vehicles. Nevertheless, the high-speed mobility of vehicles leads to frequent changes in channel state information and distances between vehicles and roadside units (RSUs), which poses challenges for low-latency perception processing. Additionally, most research overlooks the impact of vehicle mobility on perception accuracy and lacks effective resource allocation strategies for multi-source perception data fusion tasks. Addressing existing research shortcomings, this paper proposes a deep reinforcement learning(DRL)-based resource allocation method. It first adopts Integrated Sensing and Communication (ISAC) technology in the same frequency band to improve spectrum efficiency and integration. Secondly, it constructs a data fusion model to enhance vehicle perception capabilities and describes the data fusion process between vehicle terminals and RSU terminals. Furthermore, this paper designs a resource allocation algorithm for multi-source perception data fusion tasks with the optimization goal of minimizing task completion delay and system average energy consumption. Considering the mobility of vehicles and the frequent changes in communication channel states, this paper transforms the constructed problem into a Markov decision process (MDP). It solves it using the Improved Dueling Twin Delayed Deep Deterministic policy gradient (ID-TD3) algorithm. Experiment results demonstrate that the proposed strategy can reasonably allocate system resources, effectively reducing task completion delay and system average energy consumption.

In Vehicular Edge Computing (VEC), a reasonable and efficient user association and resource allocation approach is a worthwhile research issue. However, most studies in Internet of Vehicles (IoV) only consider vehicle mobility and IoV communication. Therefore, we propose a user association and resource allocation strategy in Integrated Sensing and Communication (ISAC)-aided VEC. Compared with existing solutions, we consider constraints such as sensing and communication interference, vehicle mobility, Road Side Unit (RSU) sensing performance, and vehicle user quality of service (QoS). By quantifying the sensing and communication performance of RSUs, we construct a user association and resource allocation model with the optimisation objective of maximising the average sensing performance and communication performance of the system. Then, combining double dueling deep Q-network (D3QN) algorithm and twin delayed deep deterministic policy gradient (TD3) algorithm, we propose a DRL algorithm based on D3QN-TD3. We represent the user association and resource allocation problem as a Markov Decision Process (MDP) and solve it using the proposed algorithm to obtain the optimal user association, channel allocation, and power allocation strategies. Experimental results show that the proposed algorithm has better performance in terms of downlink transmission rate, radar sensing mutual information, system utility, and task completion rate.

Vehicular edge computing (VEC) has emerged to address the increasing demands on wireless networks posed by massive data and diverse applications in intelligent vehicular services. However, challenges such as low spectrum utilization due to massive sensor deployment, signal degradation from high-speed mobility, and computational resource allocation issues hinder the real-time and secure operation of intelligent vehicles. Therefore, we propose a resource allocation optimization method for VEC based on Integrated Sensing and Communication (ISAC) and Orthogonal Time Frequency Space (OTFS) technologies. Specifically, OTFS is leveraged to multiplex roadside unit (RSU) radar resources, improving spectrum efficiency. We develop comprehensive models for communication, vehicle mobility, sensing, delay, energy consumption, and formulate a delay-minimization resource allocation problem. The problem is modeled as a Markov Decision Process and solved with an improved Twin Delayed Deep Deterministic Policy Gradient (TD3) algorithm, which incorporates prioritized experience sampling and dynamic parameter update to accelerate training and enhance agent-environment interaction. Extensive simulations are conducted in a VEC environment, where the proposed algorithm is compared with DDQN, MADDPG, and MRL-DDPG. The results demonstrate that our method effectively mitigates the impact of vehicle mobility on signal transmission and significantly reduces task completion delay compared with existing algorithms.

In congested road environments, the spectrum resources available for Roadside Units (RSUs) are often insufficient to meet the communication needs of a large number of users simultaneously. To address this, Unmanned Aerial Vehicles (UAVs) can be deployed to supplement spectrum resources temporarily. This article proposes a UAV-assisted Vehicle Edge Computing (VEC) system, integrating UAVs to enhance RSU capabilities in congested scenarios. Traditional spectrum sensing techniques, however, struggle to autonomously monitor vehicular movements and maintain stable spectrum performance. To overcome this, we introduce radar sensing devices into the RSUs to improve perception accuracy and consistency. The integration of radar sensors, while beneficial, creates additional competition for limited system resources. We, therefore, formulate the resource allocation problem considering computation delay, communication rate, and perception data, constrained by spectrum resources, offloading decisions, and time-slot allocations. The problem is modeled as a Markov Decision Process (MDP), and we propose an Improved Multi-Agent Proximal Policy Optimization (IMAPPO) algorithm to optimize resource allocation under these constraints. The experimental results show that compared to baseline algorithms such as A3C, our proposed algorithm reduces the average task processing delay by 15.53%, increases the radar estimation mutual information (MI) by 9.52%, and improves the task completion rate by 4.1%.

With the proliferation of mobile applications, the explosion of mobile data traffic imposes a significant burden on backhaul links with limited capacity in heterogeneous cellular networks (HetNets). To alleviate this challenge, content caching based on popularity at Small Base Stations (SBSs) has emerged as a promising solution. However, accurately predicting the file popularity profile for SBSs remains a key challenge due to variations in content characteristics and user preferences. Moreover, factors such as content size and the length of time slots (that is, the time duration of the update cycle for SBSs) critically impact the performance of caching schemes with limited storage capacity. In this paper, arealism-orientedintelligent caching (RETINA) is proposed to address the problem of content caching with unknown file popularity profiles, considering varying content sizes and time slots lengths. Our simulation results demonstrate that RETINA can significantly enhance the cache hit rate by 4%–12% compared to existing content caching schemes.

UAV-assisted VEC can provide content caching services for vehicles by flying close to the vehicles for vehicle's QoS. However, in real-world scenarios with traffic congestion, due to the battery capacity and cache space limitations of UAVs, low content response speed and high response latency may occur. Based on this, we proposed a dynamic energy consumption-based content caching strategy in UAV-assisted VEC. We use the PSO algorithm to solve the problem and obtain the optimal UAV deployment location. For content caching, we construct a content caching model by considering UAV deployment, vehicle user preference, UAV cache capacity, and UAV energy consumption with the goal of minimizing content request latency. In addition, we propose an IAFSA-based content caching strategy. We reduce the solution space of the fish swarm algorithm, decrease the number of caching decisions, and improve the convergence performance of AFSA by employing dynamic horizons and step sizes. Experimental results show that the proposed IAFSA effectively reduces the average content request latency of the vehicle, improves the cache hit rate, and reduces the number of content return trips. Particularly, the proposed strategy reduces the average content request latency by more than 9.84% compared to the baseline algorithm.

This paper proposes multipath routing for LEO satellite networks with a priority path and improved Bhandari mechanism (PP-IBM). PP-IBM optimizes the Bhandari algorithm and considers delay, jitter, and packet loss rate, introducing the priority mechanism and redefining the priority function. Simulation results show that compared with the other three algorithms, discrete-time dynamic virtual topology routing (DT-DVTR), optimized DT-DVTR, and time-based graph management (TGM), PP-IBM generates a minimum reduction of about 41% in routing overhead, the packet loss rate is only about 0.02%, and the jitter is as low as about 0.1ms.

In many brain network studies, brain functional connectivity data is extracted from neuroimaging data and then used for disease prediction. For now, brain disease data not only has a small sample but also has the problem of high dimensional and nonlinear. Therefore, deep clustering on brain functional connectivity data is very challenging. To solve these problems, we propose a Soft-orthogonal Constrained Dual-stream Encoder with Self-supervised clustering network (SSCDE), which consists of a pretext task and downstream task, which can fully mine the effective information in brain disease data. In the pretext task, we use two brain disease data under the same category to do cross-domain learning to obtain effective information from the same dataset. In the downstream task, to reduce redundancy and avoid negative coding, we propose a soft-orthogonal constrained dual-stream encoder to encode features separately. At the same time, we use the pseudo labels given by the pretext task as prior information for self-supervised learning. We conduct validation on different brain disease recognition tasks, and the result have proved that the proposed framework has achieved good performance compared with the unsupervised clustering analysis algorithms. To our knowledge, this is the first cross-domain assisted recognition study on brain functional connectivity data. The code is available at https://github.com/hulu88/SSCDE .

Edge caching is a promising technology to reduce backhaul strain and content access delay in Internet of Vehicles (IoV). It precaches frequently used contents close to vehicles through intermediate roadside units. Previous edge caching works often assume that content popularity is known in advance or obeys simplified models. However, such assumptions are unrealistic, as content popularity varies with uncertain spatial-temporal traffic demands in IoVs. Federated learning (FL) enables vehicles to predict popular content with distributed training. It preserves the training data remain local, thereby addressing privacy concerns and communication resource shortages. This article investigates a mobility-aware edge caching strategy by exploiting asynchronous FL and deep reinforcement learning (DRL). We first implement a novel asynchronous FL framework for local updates and global aggregation of stacked autoencoder (SAE) models. Then, utilizing the latent features extracted by the trained SAE model, we adopt a hybrid filtering model for predicting and recommending popular content. Furthermore, we explore intelligent caching decisions after content prediction. Based on the formulated Markov decision process (MDP) problem, we propose a DRL-based solution, and adopt neural network-based parameter approximations for the curse of dimensionality in RL. Extensive simulations are conducted based on real-world data trajectory. Especially, our proposed method outperforms federated averaging, least recently used, and NoDRL, and the edge hit rate is improved by roughly 6%, 21%, and 15%, respectively, when the cache capacity reaches 350 MB.

Integrated sensing and communication (ISAC) is envisioned as a key pillar for enabling the upcoming sixth generation (6G) communication systems, requiring not only reliable communication functionalities but also highly accurate environmental sensing capabilities. In this paper, we design a novel networked ISAC framework to explore the collaboration among multiple users for environmental sensing. Specifically, multiple users can serve as powerful sensors, capturing back scattered signals from a target at various angles to facilitate reliable computational imaging. Centralized sensing approaches are extremely sensitive to the capability of the leader node because it requires the leader node to process the signals sent by all the users. To this end, we propose a two-step distributed cooperative sensing algorithm that allows low-dimensional intermediate estimate exchange among neighboring users, thus eliminating the reliance on the centralized leader node and improving the robustness of sensing. This way, multiple users can cooperatively sense a target by exploiting the block-wise environment sparsity and the interference cancellation technique. Furthermore, we analyze the mean square error of the proposed distributed algorithm as a networked sensing performance metric and propose a beamforming design for the proposed network ISAC scheme to maximize the networked sensing accuracy and communication performance subject to a transmit power constraint. Simulation results validate the effectiveness of the proposed algorithm compared with the state-of-the-art algorithms.

In the digital divide regions, the edge computing can improve the performance of application services for the Internet of Things (IoT) devices. However, the lagging of information and communication technology (ICT) results in congested access spectrum and imbalanced computational load. Moreover, the mobility of IoT devices further exacerbates the fluctuating quality of communication links and the frequent changing of access positions. So, how to realize the reliable service requirements of devices in a heterogeneous environment with multiscale constraints should be considered appropriately and comprehensively. In this article, we model a relay-assisted multiaccess edge computing (MEC) framework, employing multihop transmission to enable the cross-domain service coverage. Under this framework, we formulate a quantitative model to characterize communication and computation processes within task migration, and derive analytical results for service latency. To improve the access resource efficiency, we adopt a joint nonorthogonal multiple access (NOMA) scheme to extend the transmission dimension, and employ proportional fairness to dynamically allocate resources. Besides, we propose a multiagent deep reinforcement learning (DRL) for optimizing the long-term task offloading scheduling, address the optimization problem of maximizing the system throughput efficiency. And we improve the action exploration and output dimensions of DRL to achieve convergence and performance enhancement. Simulation and analytical results show that our proposed algorithm outperforms the comparison algorithms in the key performance indicators.

Federated Learning (FL), which has been employed to train machine learning models on the data with a distributed manner, could enhance the immersive user experience for the human-centric metaverse. However, it’s challenging to train machine learning models accurately and promptly with FL for the human-centric metaverse due to massive data communication and user unreliability. User experience could be negatively affected by using low-quality machine learning models for human-centric metaverse, e.g., it cannot scrutinize and arrive at decisions accurately and timely. To resolve this pressing issue, we propose MetaFul a federated unlearning solution which reduces the negative influences of low-quality data with no data transmission by removing low-quality training models at the server side. To be specific, MetaFul includes three main components. (i) Low-throughput federated learning (LT-FL) addresses the issue of large model transmission in FL by decreasing the dimension and the number of transmitted model parameters. (ii) Loss-based model quality assessment (LM-QA) utilizes the model loss generated in LT-FL to estimate user data quality. (iii) Non-communicative federated unlearning (NC-FUL) revokes the low-quality data impact on the FL model with careful designed federated unlearning at the server side. Both LM-QA and NC-FUL have no communications with clients. Finally, extensive evaluations are conducted to show MetaFul could improve the model accuracy by at least 2.5% and decrease the user perception time by at least 19.3% in human-centric metaverse compared to benchmarks.

Common feature representation in optical and synthetic aperture radar (SAR) image registration is one of the most challenging tasks due to the significant geometric and radiometric differences. This letter proposed aProminent structure-guided feature (PSGF)representation for SAR and optical image registration. Firstly, the prominent structure of the image is highlighted based on windowed inherent variations, which is conducive to identifying more accurate and reliable corresponding points. Secondly, the maximum response index filter banks are proposed to extract structure features with multi-orientation convolution results. Then the structure feature-guided representation generated from this filtering map is quantized in histograms. Finally, the descriptor with radiation invariance is employed for feature matching, enabling automatic image registration with high accuracy. Comparative analysis with state-of-the-art methods on diverse terrain data demonstrates the superiority of the proposed PSGF method for SAR and optical image registration.

How to properly involve text characteristics like multi-scale, arbitrary direction, length aspect ratio, into detection network design has become a hot topic in computer vision. Feature Pyramid Network (FPN) is a typical method to achieve robust text detection, where its low-level and high-level feature map retains spatial structure and global semantic information, respectively. However, its strict hierarchical structure fails to fuse low-level and high-level information to improve the distinguish ability of feature map. To address this problem, we propose a novel feature fusion pyramid network for end-to-end scene text detection by fusing multi-modal information. By dividing pyramid structure into high-level and low-level layers, channel and spatial attention modules are adopted to enhance high-level and low-level feature representation by encoding channel and spatial-wise context information, respectively. In order to reduce information loss by layer transmission, a special residual network is designed to achieve short-cut between high-level and low-level features, so as to realize multi-modal feature fusion. Experiments show the precision and recall of the proposed method on ICDAR2015, ICDAR2017-MLT, and MSRA-TD500 datasets reach 88.7%/82.1%, 77.0%/60.3%, and 85.3%/74.8%, respectively.

Limited data usually cause deep neural networks to hold poor performance after training, and many generative models are proposed to synthesize data to improve the performance of models. However, existing models ignore capturing the small defect details (e.g., features and locations), resulting in that most models cannot augment the Defect Location Sensitive Data (DLS data) in which the ratio of object size to the image size is small (e.g., 20%) and the locations of the defects are only on the object. In this paper, we propose a new augmentation model, named Defect Location Sensitive data augmentation GAN (DLS-GAN), to address DLS data augmentation problem. First, we modify the vanilla generator with two Encoder-Decoder models, and view the limited masked-images masked by labeling the defect-free pixels while remaining the defect pixels in defect images and many defect-free images as the input of the two models. The extracted feature map from the first Encoder-Decoder model provides the defect features and location information; the second one extracts the features of defect-free images, and integrates the two different features with a designed Defect Feature Transfer Module to synthesize images with desired defects. Second, we employ two discriminators to estimate the scores of both distribution matching degree and defect similarity between real data and generated ones. With the two modifications, we design a new loss function, and then prove that it makes our model get converged. Last, we conduct extensive experiments to demonstrate the significant performance improvement and generalizability of DLS-GAN on different types of DLS datasets. The experimental results show that our DLS-GAN outperforms the SOTA generative models in terms of synthesizing high quality images with desired defects.Note to Practitioners—Automated defect image detectors play an important role in the field of automated manufacturing. Training a detector with superior detection performance usually requires a large number of samples. However, it is difficult to collect many defect samples in practice. Although existing generative methods can synthesize realistic-like images, they cannot generate the Defect Location Sensitive Data (DLS Data) which refer to the samples that the defects appear at the specific locations in product objects, resulting in the synthesized images invalid. This paper proposes a new defect image generation model called DLS-GAN to address this problem, and validates its performance in different real-world industrial datasets ranging from DLS Data to Non-DLS Data. Such generated images can be adopted as useful resources for improving the detection performance of automated detector.

Deep learning methods have achieved great success in medical image analysis domain. However, most of them suffer from slow convergency and high computing cost, which prevents their further widely usage in practical scenarios. Moreover, it has been proved that exploring and embedding context knowledge in deep network can significantly improve accuracy. To emphasize these tips, we present CDT-CAD, i.e., context-aware deformable transformers for end-to-end chest abnormality detection on X-Ray images. CDT-CAD firstly constructs an iterative context-aware feature extractor, which not only enlarges receptive fields to encode multi-scale context information via dilated context encoding blocks, but also captures unique and scalable feature variation patterns in wavelet frequency domain via frequency pooling blocks. Afterwards, a deformable transformer detector on the extracted context features is built to accurately classify disease categories and locate regions, where a small set of key points are sampled, thus leading the detector to focus on informative feature subspace and accelerate convergence speed. Through comparative experiments on Vinbig Chest and Chest Det 10 Datasets, CDT-CAD demonstrates its effectiveness in recognizing chest abnormities and outperforms 1.4% and 6.0% than the existing methods in$AP_{5}0$and$AR$on VinBig dateset, and 0.9% and 2.1% on Chest Det-10 dataset, respectively.

As the computing resources and the battery capacity of mobile devices are usually limited, it is a feasible solution to offload the computation-intensive tasks generated by mobile devices to edge servers (ESs) in mobile edge computing (MEC) . In this article, we study the multi-user multi-server task offloading problem in MEC systems, where all users compete for the limited communication resources and computing resources. We formulate the offloading problem with the goal of minimizing the cost of the users and maximizing the profits of the ESs. We propose a hierarchical EETORP (Economic and Efficient Task Offloading and Resource Purchasing) framework that includes a two-stage joint optimization process. Then we prove that the problem is NP-complete. For the first stage, we formulate the offloading problem as a multi-channel access game (MCA-Game) and prove theoretically the existence of at least one Nash equilibrium strategy in MCA-Game. Next, we propose a game-based multi-channel access (GMCA) algorithm to obtain the Nash equilibrium strategy and analyze the performance guarantee of the obtained offloading strategy in the worst case. For the second stage, we model the computing resource allocation between the users and ESs by Stackelberg game theory, and reformulate the problem as a resource pricing and purchasing game (PAP-Game). We prove theoretically the property of incentive compatibility and the existence of Stackelberg equilibrium. A game-based pricing and purchasing (GPAP) algorithm is proposed. Finally, a series of both parameter analysis and comparison experiments are carried out, which validate the convergence and effectiveness of the GMCA algorithm and GPAP algorithm.

Edge intelligence has emerged as a promising paradigm to accelerate DNN inference by model partitioning, which is particularly useful for intelligent scenarios that demand high accuracy and low latency. However, the dynamic nature of the edge environment and the diversity of end devices pose a significant challenge for DNN model partitioning strategies. Meanwhile, limited resources of the edge server make it difficult to manage resource allocation efficiently among multiple devices. In addition, most of the existing studies disregard the different service requirements of the DNN inference tasks, such as its high accuracy-sensitive or high latency-sensitive. To address these challenges, we propose a Multi-Compression Scale DNN Inference Acceleration (MCIA) based on cloud-edge-end collaboration. We model this problem as a mixed-integer multi-dimensional optimization problem, jointly optimizing the DNN model version choice, the partitioning choice, and the allocation of computational and bandwidth resources to maximize the tradeoff between inference accuracy and latency depending on the property of the tasks. Initially, we train multiple versions of DNN inference models with different compression scales in the cloud, and deploy them to end devices and edge server. Next, a deep reinforcement learning-based algorithm is developed for joint decision making of adaptive collaborative inference and resource allocation based on the current multi-compression scale models and the task property. Experimental results show that MCIA can adapt to heterogeneous devices and dynamic networks, and has superior performance compared with other methods.

In vehicular networks, some edge servers may not function properly due to the time-varying load condition and the uneven computing resource distribution, resulting in a low quality of caching services. To overcome this challenge, we develop a Vehicular dew computing (VDC) architecture for the first time by combining dew computing with vehicular networks, which can achieve wireless communication between vehicles in a resource-constrained environment. Consequently, it is crucial to develop an adaptive caching scheme that empowers vehicles to form efficient cooperation in VDC. In this paper, we propose an intelligent caching scheme based on VDC architecture, which includes two parts. First, to meet the dynamic nature of VDC, a spatiotemporal vehicle clustering algorithm is proposed to establish adaptive cooperation to assist content caching for vehicles. Second, the multi-armed bandit algorithm is employed to select suitable content for caching in vehicles based on real-time file popularity, and a model is established to dynamically update each vehicle’s request preferences. Extensive experiments are conducted to demonstrate that the proposed scheme has excellent performance in terms of cluster head stability and cache hit rate.

The Internet of Things (IoT) mainly consists of a large number of Internet-connected devices. The proliferation of untrusted third-party IoT applications has led to an increase in IoT-based malware attacks. In addition, it is infeasible for the IoT devices to support the sophisticated detection systems due to the restricted resources. Edge computing is considered to be promising. It provides solutions to the data security and privacy leakage brought by untrusted third-party IoT applications. In this article, an intelligent trusted and secure edge computing (ITEC) system is proposed for IoT malware detection. In this system, a signature-based preidentification mechanism is built for matching and identifying the malicious behaviors of untrusted third-party IoT applications. A delay strategy is then embedded into the risk detection engine in order to “buy time” for threat analysis and rate-limit the impact of suspicious third-party IoT applications in the system. We conduct extensive experiments to verify the effectiveness of the ITEC system and show that we can achieve accuracies of up to 98.52%.

Owing to the fast developments of economics, a lot of devices and objects have been connected and have formed the Internet of Things (IoT). Visual sensors have been applied in vehicle navigation, traffic situational awareness, and traffic safety management. However, the particles in the air degrade the imaging quality, which affects the performance of vehicle navigation, traffic situational awareness, and traffic safety management. Deep-learning-based dehazing methods were proposed to address this issue. However, these methods are trained with simulated hazy images and cannot generalize to natural haze images well. To address the domain shift problem, some methods resort to zero-shot learning or domain adaption to boost the generalization of the model on natural haze images. However, the relevance between dehazed results and clean images is ignored by zero-shot dehazing methods. Domain-adaption-based dehazing methods ignore the relationship between the dehazed results and the hazy images. To overcome these issues, a generative adversarial and self-supervised dehazing network is introduced to boost the dehazing performance on real haze images. First, generative adversarial is employed to construct the relevance between dehazed results and haze-free images, which can boost the natural appearance of dehazed results. Second, self-supervised learning is employed to construct the relevance between the dehazed results and hazy images, which can restrict the solution space of dehazing. To show the effectiveness of the proposed model, we conduct extensive experiments on real and simulated haze images. Compared with state-of-the-art methods, the proposed model achieves state-of-the-art dehazing performance.

Although collaborative 3D perception has successfully improved detection performance by sharing LIDAR information among multiple agents, its impact under adverse weather is under poor investigation. It is non-trivial to reduce the noise effect in the multi-agent system, as each agent may generate defective feature representations with aleatoric uncertainty, and such uncertainty will be further amplified in the collaborative stage due to deterministic collaboration models. To mitigate the negative effects of weather noise on the collaborative framework, we proposed a method called Co-Denoising, which incorporates a two-stage denoising approach within the intermediate collaborative framework. In our method, a sampling-based noise filtering is first performed at each agent to make a coarse denoising. Then, during the collaboration stage, the global feature representations are expanded through Bayesian neural networks to improve the robustness against environmental noise. The extensive experiments on sunny and rainy datasets have indicated the proposed collaborative perception method can significantly reduce performance degradation under adverse weather.

It is inevitable to recognize objects in adverse weather conditions where the uncertainty of contour areas is increased. Although some multi-task learning frameworks have gained from the directional supervision between boundary detection and semantic segmentation, the interaction between those two tasks is poorly investigated. Moreover, the performance of the contour detection is expected to degrade under foggy scenarios, because the auxiliary task also has no benefits from the main task. To address the potential risk in intelligent transportation systems, this paper proposes a mutual learning framework, named CE-MGN (Confidence-Enhanced Mutual Graph Network), to propagate confidence through continuous interaction between different tasks rather than only focusing on the accuracy of the main task. The CE-MGN performs an end-to-end training paradigm and jointly learns two tasks, contour detection and semantic segmentation, through pairwise confidence-enhancement mechanism. Moreover, the task interaction is converted into graph space to further relieve the information loss during the feature aggregation in Euclidean space. Such a framework is capable to improve the robustness of respective tasks because of the encouragement from its peer task. Extensive experiments show that our CE-MGN achieved mean IoU scores of 79.35% and 79.03% on CityScapes and Foggy CityScapes datasets, respectively. Besides, our models have a stable performance on different weather severity, where the performance fluctuation is less than 1%.

The outbreak of COVID-19 has caused a serious impact on the traditional logistics industry. Considering that the truck-drone collaborative delivery system can both reduce the risk of COVID-19 propagation and deliver supplies in a cost effective and timely manner, this paper introduces the Multiple visits Travelling Salesman Problem with Multiple Heterogeneous Drones (MTSP-MHD). The model allows a truck to carry a fleet of heterogeneous multi-visit drones for cooperative deliveries, where the drones are capable of delivering to multiple customers on a single route and the flight is restricted by energy consumption and payload constraints. To solve MTSP-MHD, we develop an approach that combines K-Means$++$clustering, Nearest neighbor search and Greedy strategies (KNG) to construct feasible solutions. Meanwhile, an Improved Artificial Bee Colony algorithm combining Metropolis acceptance criterion of Simulated Annealing, Tabu list of Tabu Search, and Elite selection strategies (IABC-MTE) is proposed to enhance the quality of solutions. Particularly, three problem-specific neighborhood operators are adopted to search for new solutions. The massive experimental results indicate that IABC-MTE achieves significant improvements over other competitors, with average objective value reductions ranging from 1.81% to 29.16% and standard deviations reduced by 0.04 to 26.44. Finally, the influencing factors of the drone fleet, the performance of different drone fleets and delivery modes are evaluated in detail.

Unmanned Aerial Vehicles (UAVs) can be employed as short-term aerial base stations or as access points for User Equipments (UEs) to communicate with other UEs effectively. However, communication links may be obstructed by buildings, leading to poor data transfer performance and significant energy consumption. Deploying Reconfigurable Intelligent Surfaces (RIS) as part of the UAV-assisted communication system proves to be an effective means to avoid building obstructions and enhance wireless information quality. However, the complexity of communication relationships in multi-UAV systems with RIS-aided communication poses a significant challenge in energy reduction. Therefore, this study investigates a new RIS-aided multi-UAV communication framework for edge computing systems. The system aims to meet the quality-of-service (QoS) for UEs while minimizing the total energy consumption. To optimize the total energy consumption of RIS-aided multi-UAV communication, the impact of communication between multiple UAVs and differences between UE clusters on that system’s performance is also considered. We introduce a Stackelberg game to deal with the communication relationship between multiple UAVs and design a K-means-based clustering algorithm to segment UEs periodically. A model-free deep reinforcement learning algorithm grounded in maximum entropy is proposed to jointly optimize UAV trajectory design, phase shift control, and power allocation to reduce energy consumption further. Experimental results indicate that the system proposed performs favorably concerning both energy consumption and throughput.

Due to the outstanding merits such as mobility, high maneuverability, and flexibility, Unmanned Aerial Vehicles (UAVs) are viable mobile power transmitters that can be rapidly deployed in geographically constrained regions. They are good candidates for supplying power to energy-limited Sensor Nodes (SNs) with Wireless Power Transfer (WPT) technology. In this paper, we investigate a UAV-enabled WPT system that transmits power to a set of SNs at unknown positions. A key challenge is how to efficiently gather the locations of SNs and design a power transfer scheme. We formulate a multi-objective optimization problem to jointly optimize these objectives: maximization of UAV's search efficiency, maximization of total harvested energy, minimization of UAV's flight energy consumption and maximization of UAV's energy utilization efficiency. To tackle these issues, we present a two-stage strategy that includes a UAV Motion Control (UMC) algorithm for obtaining the coordinates of SNs and a Dynamic Genetic Clustering (DGC) algorithm for power transfer via grouping SNs into clusters. First, the UMC algorithm enables the UAV to autonomously control its own motion and conduct target search missions. The objective is to make the energy-restricted UAV find as many SNs as feasible without any apriori knowledge of their information. Second, the DGC algorithm is used to optimize the energy consumption of the UAV by combining a genetic clustering algorithm with a dynamic clustering strategy to maximize the amount of energy harvested by SNs and the energy utilization efficiency of the UAV. Finally, experimental results show that the proposed algorithms outperform their counterparts.

Vehicular Edge Computing (VEC) is the transportation version of Mobile Edge Computing (MEC). In VEC, task offloading enables vehicles to offload computing tasks to nearby Roadside Units (RSUs), thereby reducing the computation cost. Recent trends in task offloading cause a proliferation of studies in academia. However, the existing offloading schemes still face many challenges, such as high-dynamic network topology, massive and complex data, dynamic scenes with high-speed vehicles and low-latency requirements. Digital Twin (DT)-based VEC is emerging as a promising solution. It monitors the state of the VEC network in real time through mappings and interactions between the physical and virtual entities. Consequently, the task offloading scheme can make more reasonable offloading decisions at the physical layer and further improve the efficiency of VEC. Above all, we propose a VEC computing offloading scheme, namely, AdaptiveSwarm Intelligent Offloading Scheme Based on Digital-Twin-Assisted PRedictionInVEC (STRIVE). The VEC network architecture is established to combines DT with an improved Generative Adversarial Network (GAN). The powerful prediction ability of GAN is used to assist in constructing DT in the pre-processing phase, reducing the size of the decision space. To adapt to the dynamic nature of VEC, we establish an adaptive model to adjust the real-time parameter under various scenarios. Then, we deploy an improveDgenetIc simulatEd annealing-baSEd particLe swarm optimization (DIESEL) algorithm to task offloading decision-making, which can provide reliable computing services for vehicles at a lower cost. The simulation results demonstrate that the proposed scheme can effectively reduce computing delay and energy consumption compared with its counterparts.

Vehicular Edge Computing (VEC) is the transportation version of Mobile Edge Computing (MEC) in road scenarios. One key technology of VEC is task offloading, which allows vehicles to send their computation tasks to the surrounding Roadside Units (RSUs) or other vehicles for execution, thereby reducing computation delay and energy consumption. However, the existing task offloading schemes still have various gaps and face challenges that should be addressed because vehicles with time-varying trajectories need to process massive data with high complexity and diversity. In this paper, a VEC-based computation offloading model is developed with consideration of data dependency of tasks. The minimization of the average response time and average energy consumption of the system is defined as a combinatorial optimization problem. To solve this problem, we propose aMobility-aware dependent taskoffloading (MESON) Scheme for urban VEC and develop a DRL-based algorithm to train the offloading strategy. To improve the training efficiency, a vehicle mobility detection algorithm is further designed to detect the communication time between vehicles and RSUs. In this way, MESON can avoid unreasonable decisions by lowering the size of the action space. Moreover, to improve the system stability and the offloading successful rate, we design a task priority determination scheme to prioritize the tasks in the waiting queue. The experimental results show that MESON is superior compared to other task offloading schemes in terms of the average response time, average system energy consumption, and offloading successful rate.

Knowledge distillation (KD) is a prevalent model compression technique in deep learning, aiming to leverage knowledge from a large teacher model to enhance the training of a smaller student model. It has found success in deploying compact deep models in intelligent applications like intelligent transportation, smart health, and distributed intelligence. Current knowledge distillation methods primarily fall into two categories: offline and online knowledge distillation. Offline methods involve a one-way distillation process, transferring unvaried knowledge from teacher to student, while online methods enable the simultaneous training of multiple peer students. However, existing knowledge distillation methods often face challenges where the student may not fully comprehend the teacher's knowledge due to model capacity gaps, and there might be knowledge incongruence among outputs of multiple students without teacher guidance. To address these issues, we propose a novel reciprocal teacher-student learning inspired by human teaching and examining through forward and feedback knowledge distillation (FFKD). Forward knowledge distillation operates offline, while feedback knowledge distillation follows an online scheme. The rationale is that feedback knowledge distillation enables the pre-trained teacher model to receive feedback from students, allowing the teacher to refine its teaching strategies accordingly. To achieve this, we introduce a new weighting constraint to gauge the extent of students' understanding of the teacher's knowledge, which is then utilized to enhance teaching strategies. Experimental results on five visual recognition datasets demonstrate that the proposed FFKD outperforms current state-of-the-art knowledge distillation methods.

In existing research on cluster head selection schemes in the Internet of Vehicles (IoV), designing a stable cluster structure poses a significant challenge. Choosing a centrally-located cluster head that can respond rapidly is crucial for meeting various requirements. To address the aforementioned challenges, this paper introduces a machine learning-based IoV cluster head selection scheme (HSCS). We introduce a new metric termed N-cycle Average Virtual Cluster Delay (XTn) for appropriate cluster head selection. To accommodate the high dynamism of vehicles, a machine learning model is integrated to predict cluster head selection metrics across different periods, and a set of cluster head selection guidelines is formulated. Experimental results demonstrate that our proposed HSCS ensures a relatively low average intra-cluster delay while maintaining a longer cluster head retention time, and it exhibits commendable robustness.

Digital twins (DT) and mobile networks have evolved forms of intelligence in Internet of Things (IoT). In this work, we consider a Digital Twin Mobile Network (DTMN) scenario with few multimedia samples. Facing challenges of knowledge extraction with few samples, stable interaction with dynamic changes of multimedia data, time and privacy saving in low-resource mobile network, we propose an edge computing and few-shot learning featured intelligent framework. Considering time-sensitive property of transmission and privacy risks of directly uploads in mobile network, we deploy edge computing to locally run networks for analysis, thus saving time to offload computing request and enhancing privacy by encrypting original data. Inspired by remarkable relationship representation of graphs, we build Graph Neural Network (GNN) in cloud to map physical mobile systems to virtual entities with DT, thus performing semantic inferences in cloud with few samples uploaded by edges. Occasionally, node features in GNN could converge to similar, non-discriminative embeddings, causing catastrophic unstable phenomena. An iterative reweight and drop structure (IRDS) is thus constructed in cloud, which nonetheless contributes stability with respect to edge uncertainty. As part of IRDS, a drop Edge&Node scheme is proposed to randomly remove certain nodes and edges, which not only enhances distinguished capability of graph neighbor patterns, but also offers data encryption with random strategy. We show one implementation case of image classification in social network, where experiments on public datasets show that our framework is effective with user-friendly advantages and significant intelligence.

The convergence of unmanned aerial vehicle (UAV)-aided mobile edge computing (MEC) networks and blockchain transforms the existing mobile networking paradigm. However, in the temporary hotspot scenario for intelligent connected vehicles (ICVs) in UAV-aided MEC networks, deploying blockchain-based services and applications in vehicles is generally impossible due to its high computational resource and storage requirements. One possible solution is to offload part of all the computational tasks to MEC servers wherever possible. Unfortunately, due to the limited availability and high mobility of the vehicles, there is still lacking simple solutions that can support low-latency and higher reliability networking services for ICVs. In this article, we study the task offloading problem of minimizing the total system latency and the optimal task offloading scheme, subject to constraints on the hover position coordinates of the UAV, the fixed bonuses, flexible transaction fees, transaction rates, mining difficulty, costs and battery energy consumption of the UAV. The problem is confirmed to be a challenging linear integer planning problem, we formulate the problem as a constrained Markov decision process. Deep Reinforcement Learning (DRL) has excellently solved sequential decision-making problems in dynamic ICVs environment, therefore, we propose a novel distributed DRL-based P-D3QN approach by using Prioritized Experience Replay strategy and the dueling double deep Q-network (D3QN) algorithm to solve the optimal task offloading policy effectively. Finally, experiment results show that compared with the benchmark scheme, the P-D3QN algorithm can bring about 26.24% latency improvement and increase about 42.26% offloading utility.

We study the problem of classifying different cooking styles, based on the recipe. The difficulty is that the same food ingredients, seasoning, and the very similar instructions result in different flavors, with different cooking styles. Existing methods have limitations: they mainly focus on homogeneous data (e.g., instruction or image), ignoring heterogeneous data (e.g., flavor compound or ingredient), which certainly hurts the classification performance. This is because collecting enough available heterogeneous data of a recipe is a non-trivial task. In this paper, we present a new heterogeneous data augmentation method to improve classification performance. Specifically, we first construct a heterogeneous recipe graph network to represent heterogeneous data, which includes four main-stream types of heterogeneous data: ingredient, flavor compound, image, and instruction. Then, we draw a sequence of augmented graphs for Semi-Supervised learning through multinomial sampling. The probability distribution of sampling depends on the Cosine distance between the nodes of graph. In this way, we name our approach as Multinomial Sampling Graph Data Augmentation (MS-GDA). Extensive experiments demonstrate that MS-GDA significantly outperforms SOTA baselines on cuisine classification and region prediction with the recipe benchmark dataset. Code is available at https://github.com/LiangzheChen/MS-GDA .

Knowledge distillation (KD) is a powerful and widely applicable technique for the compression of deep learning models. The main idea of knowledge distillation is to transfer knowledge from a large teacher model to a small student model, where the attention mechanism has been intensively explored in regard to its great flexibility for managing different teacher-student architectures. However, existing attention-based methods usually transfer similar attention knowledge from the intermediate layers of deep neural networks, leaving the hierarchical structure of deep representation learning poorly investigated for knowledge distillation. In this paper, we propose a hierarchical multi-attention transfer framework (HMAT) , where different types of attention are utilized to transfer the knowledge at different levels of deep representation learning for knowledge distillation. Specifically, position-based and channel-based attention knowledge characterize the knowledge from low-level and high-level feature representations, respectively, and activation-based attention knowledge characterize the knowledge from both mid-level and high-level feature representations. Extensive experiments on three popular visual recognition tasks, image classification, image retrieval, and object detection, demonstrate that the proposed hierarchical multi-attention transfer or HMAT significantly outperforms recent state-of-the-art KD methods.

No abstract available.

Generative adversarial networks (GANs)-based person re-identification (re-id) schemes provide potential ways to augment data in practical applications. However, existing solutions perform poorly because of the separation of data generation and re-id training and a lack of diverse data in real-world scenarios. In this paper, a person re-id model (IDGAN) based on semantic map guided identity transfer GAN is proposed to improve the person re-id performance. With the aid of the semantic map, IDGAN generates pedestrian images with varying poses, perspectives, and backgrounds efficiently and accurately, improving the diversity of training data. To increase the visual realism, IDGAN utilizes a gradient augmentation method based on local quality attention to refine the generated image locally. Then, a two-stage joint training framework is employed to allow the GAN and the person re-id network to learn from each other to better use the generated data. Detailed experimental results demonstrate that, compared with the existing state-of-the-art methods, IDGAN is capable of producing high-quality images and significantly enhancing re-id performance, with the FID of generated images on the Market-1501 dataset being reduced by 1.15, and mAP on the Market-1501 and DukeMTMC-reID datasets being increased by 3.3% and 2.6%, respectively.

In vehicular edge computing, Unmanned Aerial Vehicles (UAVs) have become a feasible solution for addressing high deployment costs faced by base stations in congested roads during peak hours. However, UAVs cannot cache all requested content due to limited storage. Hence, we propose a content caching strategy based on user preference predictions. To address resource consumption and user privacy concerns during the training process, we propose a user preference prediction model based on hierarchical federated learning training. Specifically, we employ a hierarchical clustering approach to partition user vehicles and UAVs into multiple clusters and utilize hierarchical federated learning to train prediction models within each cluster. Furthermore, to tackle the joint optimization problem of content caching and bandwidth allocation, we propose I-MADDPG, an improved multi-agent deep deterministic policy gradient algorithm. It determines the next continuous action based on the reward value at the current moment and the average reward value in the iteration period as reference parameters. The experimental results demonstrate that the proposed algorithm has significantly enhanced training efficiency compared to the baselines. Additionally, it has improved cache hit rate and reduced content request delay through effective resource allocation.

Wireless power communication networks(WPCN) can be used in disaster scenarios to provide reliable communication and power supply, helping to improve disaster response efforts. Using reconfigurable intelligent surface(RIS) to assist WPCN can improve signal quality and extend the transmission distance where traditional infrastructure is damaged or unavailable. However, it is important to transmit an emergency response in disaster scenarios as soon as possible, which is little consideration before. In this paper, we propose a time allocation scheme whose key idea is to find the minimum time for disaster scenario information dissemination while optimizing the time threshold for downlink energy harvest to confirm uplink information transmit requirements. We propose a new optimization problem of minimizing the transmit time duration by jointly optimizing the RIS shifts, downlink time for energy havest, uplink time and power for information transmission, subject to the constraints on harvesting power at user, the phase shift module 1, and the minimum Quality of Service(QoS) constraints. We find that our proposed optimization problem is multi variables and non-convex. To solve this, we apply alternating optimization to divide it into two sub-problems. Then Successive Convex Approximation and penalty-based algorithm are introduced to solve the sub-problems, respectively. Simulation results show that we can find the least time for information transmission in disaster scenarios.

As a new network technology, mobile-edge computing (MEC) combined with the Internet of Vehicles (IoV) can effectively improve the efficiency of task computing and offloading. However, the power of edge computing will be severely limited to the areas with poor MEC server coverage. Furthermore, there are a number of peripheral vehicles with temporarily idle computing resources on the road, so how to put the resources of these vehicles into use becomes the primary issue to be considered. In this article, a distributed multihop task offloading decision model for task execution efficiency is developed, which mainly consists of two parts: 1) a candidate vehicle selection mechanism for screening the neighboring vehicles that can participate in offloading and 2) a task offloading decision algorithm for obtaining the task offloading solution. Considering the impact of different hop and wireless communication ranges on communication ranges on task completion in a generic scenario, we introduce the hop count$k$and select the neighboring vehicles in the$k$-hop wireless communication range as the candidate vehicles. Then, the problem of offloading is modeled as a generalized allocation model with constraints which is solved by the greedy algorithm and discrete bat algorithm, respectively. The results show that compared with the scheme in which the task vehicle randomly selects the neighboring vehicles to offload and the scheme that all tasks are completed locally, the offloading scheme in which all tasks are completed under the greedy algorithm or bat-based algorithm has advantages in time delay performance in terms of different task number, task required computation power, and task size environment. Besides, this article also explores the influence of hop count$k$on the results when selecting candidate vehicles from the neighboring vehicles within the range of$k$hop. The results show that the increase of$k$will also increase the number of candidate vehicles, which makes the time delay lower. Under the parameters set in this article, the time delay required for the greedy algorithm offloading scheme to complete all tasks is a lower bound on the time delay of the bat algorithm scheme. The greedy algorithm scheme reduces latency by 0.2–2.4 s compared to the scheme where tasks are all completed locally, and it reduces latency by 0.16–2.3 s compared to the random offloading scheme.

Data-driven ontology-based knowledge (OK) presentation and computational linguistics for evolving semantic Asian social networks (ASNs) can make one of the most important platforms that provide robust and real-time data mapping in massive access across the heterogeneous big data sources in the web that is named OK-ASN. It benefits from computational intelligence, web-of-things (WoT) architecture, semantic features, statistical learning and pattern recognition, database management, computer vision, cyber-security, and language processing. OK-ASN is a critical strategy for WoT big data mining and enterprises from social media to medical and industrial sectors.

How to accurately understand low-resource languages is the core of the task-oriented human-computer dialogue system. Language understanding consists of two sub-tasks, i.e., intent detection and slot filling. Intent detection still faces challenges due to semantic ambiguity and implicit intentions with users’ input. Moreover, separately modeling intent detection and slot filling significantly decrease the correctness and relevance between questions and answers. To address these issues, we propose a joint intent detection method using asynchronous training strategy. The proposed method firstly encodes local text information extracted by CNN and relationship information among words emphasized by attention structure. Later, a joint intent detection model with asynchronous training strategy is proposed by either fusing hidden states of intent detection and slot filling layers, or adopting the key information to fine-tune the whole network, greatly increasing the relevance of intent detection and slot filling subtasks. The accuracy achieved by the proposed method tested on an open-source airline travel dataset and a self-collected electricity service dataset, i.e., ATIS and ECSF, are 97.49% and 89.68%, respectively, which proves the effectiveness of joint learning and asynchronous training.

Modern Healthcare cyberphysical systems have begun to rely more and more on distributed AI leveraging the power of Federated Learning (FL). Its ability to train Machine Learning (ML) and Deep Learning (DL) models for the wide variety of medical fields, while at the same time fortifying the privacy of the sensitive information that are present in the medical sector, makes the FL technology a necessary tool in modern health and medical systems. Unfortunately, due to the polymorphy of distributed data and the shortcomings of distributed learning, the local training of Federated models sometimes proves inadequate and thus negatively imposes the federated learning optimization process and in extend in the subsequent performance of the rest Federated models. Badly trained models can cause dire implications in the healthcare field due to their critical nature. This work strives to solve this problem by applying a post-processing pipeline to models used by FL. In particular, the proposed work ranks the model by finding how fair they are by discovering and inspecting micro-Manifolds that cluster each neural model's latent knowledge. The produced work applies a completely unsupervised both model and data agnostic methodology that can be leveraged for general model fairness discovery. The proposed methodology is tested against a variety of benchmark DL architectures and in the FL environment, showing an average 8.75% increase in Federated model accuracy in comparison with similar work.

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

Although vertical federated learning (VFL) has become a new paradigm of distributed machine learning for emerging multiparty joint modeling applications, how to effectively incentivize self-conscious clients to actively and reliably contribute to collaborative learning in VFL has become a critical issue. Existing efforts are inadequate to address this issue since the training sample size needs to be unified before model training in VFL. To this end, selfish clients should unconditionally and honestly declare their private information, such as model training costs and benefits. However, such an assumption is unrealistic. In this article, we develop the first Truthful incEntive mechAnism for VFL,$\mathbb {TEA}$, to handle both information self-disclosure and social utility maximization. Specifically, we design a transfer payment rule via internalizing externalities, which bundles the clients’ utilities with the social utility, making truthful reporting by clients be a Nash equilibrium. Theoretically, we prove that$\mathbb {TEA}$can achieve truthfulness and social utility maximization, as well as budget balance (BB) or individual rationality (IR). On this basis, we further design a sample size decision rule via linear programming (LP) relaxation to meet the requirements of different scenarios. Finally, extensive experiments on synthetic and real-world datasets validate the theoretical properties of$\mathbb {TEA}$and demonstrate its superiority compared with the state-of-the-art.

Facial Expression Recognition (FER) in the wild poses significant challenges due to realistic occlusions, illumination, scale, and head pose variations of the facial images. In this article, we propose an Edge-AI-driven framework for FER. On the algorithms aspect, we propose two attention modules, Arbitrary-oriented Spatial Pooling (ASP) and Scalable Frequency Pooling (SFP), for effective feature extraction to improve classification accuracy. On the systems aspect, we propose an edge-cloud joint inference architecture for FER to achieve low-latency inference, consisting of a lightweight backbone network running on the edge device, and two optional attention modules partially offloaded to the cloud. Performance evaluation demonstrates that our approach achieves a good balance between classification accuracy and inference latency.

Falsified faces generated by DeepFake are severe threats to our community. Many smart systems in Industry 4.0, such as electronic payments and identity verification, rely on bioinformation authentication. These applications may compromise with forgeries generated by DeepFake. Notwithstanding many promising results on DeepFake forensics have been reported recently, we are now facing new security challenges brought by antiforensics attacks. With adversarial perturbations injected by antiforensics algorithms, falsified faces could masquerade themselves to disrupt forensics detectors as well as industrial applications. Therefore, to secure biometric data, in particular facial information, we propose a countermeasure against the attacks of DeepFake antiforensics. The proposed model features dual channels and multiple supervisors to capture biological attributes from manifold aspects. After training, the proposed method can purify antiforensics images by eliminating adversarial perturbations. With experimental evaluations, we show that purified faces are highly distinguishable from real ones. The proposed method is justified as a reliable defense tool for protecting facial bioinformation against antiforensics amid Industry 4.0.

Recently, model compression has been widely used for the deployment of cumbersome deep models on resource-limited edge devices in the performance-demanding industrial Internet of Things (IoT) scenarios. As a simple yet effective model compression technique, knowledge distillation (KD) aims to transfer the knowledge (e.g., sample relationships as the relational knowledge) from a large teacher model to a small student model. However, existing relational KD methods usually build sample correlations directly from the feature maps at a certain middle layer in deep neural networks, which tends to overfit the feature maps of the teacher model and fails to address the most important sample regions. Inspired by this, we argue that the characteristics of important regions are of great importance, and thus, introduce attention maps to construct sample correlations for knowledge distillation. Specifically, with attention maps from multiple middle layers, attention-based sample correlations are newly built upon the most informative sample regions, and can be used as an effective and novel relational knowledge for knowledge distillation. We refer to the proposed method as multilevel attention-based sample correlations for knowledge distillation (or MASCKD). We perform extensive experiments on popular KD datasets for image classification, image retrieval, and person reidentification, where the experimental results demonstrate the effectiveness of the proposed method for relational KD.

Despite the rapid growth of successful examples of Federated Learning (FL), it faces the heterogeneity of data, models, and devices in emerging applications of Industrial Internet of Things (IIoT). Existing efforts mainly focus on training multiple personalized models by adopting a global, cluster, or pairwise fashion. However, the global collaboration does not work well in case of the non-IID data distribution, cluster collaboration is often inefficient due to the single cluster pattern and high computational cost, and pairwise collaboration incurs the limitations of collaboration scope and communication efficiency. To address the problems, we propose a novel personalized FL (PFL) framework with the game-theoretic insights, called group collaboration, to overcome the shortcomings of status quo. Specifically, we first formulate the group collaboration in PFL as a multileader multifollower Stackelberg game, and then develop an$\epsilon$-better response to efficiently characterize its unique equilibrium through cautiously proposing a potential function. Since the existing equilibrium may not be optimal, we further design a Robin Hood mechanism by using the idea of transferable utility to improve the performance of the training model. Meanwhile, we also prove that the new mechanism is sustainable and can converge to a stable state with an upper bound of the training loss. Last, extensive experiments on a simulated dataset and four real-world datasets demonstrate the superiority of our proposed approach compared to the state of the art.

There exists a rising concern on security of healthcare data and service. Even small lost, stolen, displaced, hacked, or communicated in personal health data could bring huge damage to patients. Therefore, we propose a novel content-aware deoxyribonucleic acid (DNA) computing system to encrypt medical images, thus guaranteeing privacy and promoting secure healthcare environment. The proposed system consists of sender and receiver to perform tasks of encryption and decryption, respectively, where both contain the same structure design, but perform opposite operations. In either sender or receiver, we design a randomly DNA encoding and a content-aware permutation and diffusion module. Considering introducing random mechanism to increase difficulty of cracking, the former module builds a random encryption rule selector in DNA encoding process by randomly mapping quantity of medical image pixels to outputs. Meanwhile, the latter module constructs a permutation sequence, which not only encodes information of pixel values, but also involves redundant correlation between adjacent pixels located in a patch. Such design brings awareness property of medical image content to greatly increase complexity in cracking by embedding semantical information for encryption. We demonstrate that the proposed system successfully improve cybersecurity of medical images against various attacks in robustness and effectiveness when transmitting data in wireless broadcasting scenarios.

Colonic adenocarcinoma is a disease severely endangering human life caused by mucosal epidermal carcinogenesis. The segmentation of potentially cancerous glands is the key in the detection and diagnosis of colonic adenocarcinoma. The appearance of cancerous tissue is different in gland segmentation in colon pathological images, and it is impossible to accurately segment the changes of glands from benign to malignant using a single network. Given these issues, a two-path gland segmentation algorithm of colon pathological image based on local semantic guidance is proposed in this paper. The improved candidate region search algorithm is adopted to expand the original image data set and generate sub-datasets sensitive to specific features. Then, the semantic feature-guided model is employed to extract the local adenocarcinoma features and acts on the backbone network together with context feature extraction based on the attention mechanism. In this way, a larger receptive field and more local feature information are obtained, the learning ability of the network to the morphological features of glands is enhanced, and the performance of automatic gland segmentation is finally improved. The algorithm is verified on Warwick Qu-Dataset. Compared with the current popular segmentation algorithms, our algorithm has good performance in Dice coefficient, F1 score, and Hausdorff distance on different types of test sets.

Automatic generation of medical reports can provide diagnostic assistance to doctors and reduce their workload. To improve the quality of the generated medical reports, injecting auxiliary information through knowledge graphs or templates into the model is widely adopted in previous methods. However, they suffer from two problems: 1) The injected external information is limited in amount and difficult to adequately meet the information needs of medical report generation in content. 2) The injected external information increases the complexity of model and is hard to be reasonably integrated into the generation process of medical reports. Therefore, we propose an Information Calibrated Transformer (ICT) to address the above issues. First, we design a Precursor-information Enhancement Module (PEM), which can effectively extract numerous inter-intra report features from the datasets as the auxiliary information without external injection. And the auxiliary information can be dynamically updated with the training process. Secondly, a combination mode, which consists of PEM and our proposed Information Calibration Attention Module (ICA), is designed and embedded into ICT. In this method, the auxiliary information extracted from PEM is flexibly injected into ICT and the increment of model parameters is small. The comprehensive evaluations validate that the ICT is not only superior to previous methods in the X-Ray datasets, IU-X-Ray and MIMIC-CXR, but also successfully be extended to a CT COVID-19 dataset COV-CTR.

Edge intelligence (EI) migrates data and artificial intelligence (AI) to the “edge” of a network, enhancing the high-bandwidth and low-latency of wireless data transmission with the multiplier effect of 5G and AI, greatly improving the edges’ processing speed. Through integrating EI and computer vision technology, video surveillance systems in ITS can improve the processing capability of traffic information, which improves traffic efficiency and ensures traffic safety. Accordingly, first, we propose an edge intelligence-based improved-YOLOv4 vehicle detection algorithm, introducing an efficient channel attention (ECA) mechanism and a high-resolution network (HRNet) to enhance vehicle detection ability. Second, an edge intelligence-based improved DeepLabv3+ image segmentation algorithm is proposed, replacing the original backbone network with MobileNetv2 and using the softpool method, thus reducing the network size while improving the segmentation accuracy. Experimental results show that our proposed model has a higher average precision (AP) and can improve vehicle detection accuracy from 82.03% to 86.22%. The mean intersection over union (mIOU) of the image segmentation model improves from 73.32% to 75.63%.

In metropolitan environments, unmanned aerial vehicles (UAVs) equipped with video surveillance equipment can monitor crowd behavior and maintain public safety. In high-traffic areas where humans are more likely to make mistakes, a smart city needs modern technology to forecast the behavior of its residents. In order to improve citywide traffic flow, urban transportation systems (UTSs) monitor and learn how people behave in crowds. Using UAVs for video surveillance in smart cities, our research describes a unique way to assess crowd condition, which expands the scope of human-vehicle interactions. Moreover, we use fuzzy logic ranking to improve the system’s ability to detect anomalies in crowds. In order to improve decision-making, a novel deep transfer learning (DTL) technique is applied to the UAV’s received frames. A 98.5% accuracy rate, satisfactory performance, and robustness to population behavior are all characteristics of the proposed integrated model. In UTSs and urban areas, our novel intelligent system analyzes human behavior based on vehicle-human interactions. In areas with low and high traffic congestion, the modified ResNet (mResNet) architecture predicts the crowd’s condition based on fuzzy logic (FLA). Through decision-making based on accurate crowd conditions, best general paths can be selected using the algorithm.

The integration of Mobile Edge Computing (MEC) and microservice architecture drives the implementation of the sustainable Internet of Vehicles (IoV). The microservice architecture enables the decomposition of a service into multiple independent, fine-grained microservices working independently. With MEC, microservices can be placed on Edge Service Providers (ESPs) dynamically, responding quickly and reducing service latency and resource consumption. However, the burgeoning of IoV leads to high computation and resource overheads, making service resource requirements an imminent issue. What’s more, due to the limited computation power of ESPs, they can only host a few services. Therefore, ESPs should judiciously decide which services to host. In this paper, we propose a Microservice-oriented Service Placement (MOSP) mechanism for MEC-enabled IoV to shorten service latency, reduce high resource consumption levels and guarantee long-term sustainability. Specifically, we formulate the service placement as an integer linear programming program, where service placement decisions are collaboratively optimized among ESPs, aiming to address spatial demand coupling, service heterogeneity, and decentralized coordination in MEC systems. MOSP comprises an upper layer to map the service requests to ESPs and a lower layer to adjust the service placement of ESPs. Evaluation results show that the microservice-oriented service deployment mechanism offers dramatic improvements in terms of resource savings, latency reduction, and service speed.

Multi-modal fusion is a promising approach to boost the autonomous driving performance and has already received a large amount of attention. Meanwhile, to increase driving reliability under distinct scenarios, it is important to handle unforeseen weather events in the training dataset, which is known as an Out-Of-Distribution (OOD) problem, for autonomous driving algorithms. In this paper, we consider those two aspects and propose an end-to-end multi-modal domain-enhanced framework, namely CrossFuser, to meet the safety orientated driving requirements. CrossFuser first integrates both image and lidar modalities to generate a robust environmental representation through conjoint mapping, elastic disentanglement, and attention mechanism. Further, the perception embedding is used to calculate corresponding waypoints by a waypoint prediction network, consisting of Gate Recurrent Units (GRUs). Finally, the final control commands are calculated by low-level control functions. We conduct experiments on the Car Learning to Act (CARLA) driving simulator involving complex weather conditions under urban scenarios, the results show that CrossFuser can outperform the state of the art.

With the remarkable technological development in cyber-physical systems, industry 4.0 has evolved by use of a significant concept named digital twin (DT). However, it is still difficult to construct a relationship between twin simulation and a real scenario considering dynamic variations, especially when dealing with small surface defect detection tasks with high performance and computation resource requirements. In this article, we aim to construct cyber-manufacturing systems to achieve a DT solution for small surface defect detection task. Focusing on DT-based solution, the proposed system consists of an Edge–Cloud architecture and a surface defect detection algorithm. Considering dynamic characteristics and real-time response requirement, Edge–Cloud architecture is built to achieve smart manufacturing by efficiently collecting, processing, analyzing, and storing data produced by factory. A deep learning–based algorithm is then constructed to detect surface defeats based on multi-modal data, i.e., imaging and depth data. Experiments show the proposed algorithm could achieve high accuracy and recall in small defeat detection task, thus constructing DT in cyber-manufacturing.

Person re-identification (Re-ID) has achieved great success in single-domain. However, it remains a challenging task to adapt a Re-ID model trained on one dataset to another one. Unsupervised domain adaption (UDA) was proposed to migrate a model from a labeled source domain to an unlabeled target domain. The main difference in the cross-domain is different background styles. Although the style transfer approach effectively reduces inter-domain gaps, it ignores the reduction of intra-class differences. Clustering-based pipelines maintain state-of-the-art performance for UDA by learning domain-independent features; however, most existing models do not sufficiently exploit the rich unlabeled samples in target domains due to unsatisfactory clustering. Thus, we propose a novel local correlation ensemble model that focuses on the diversity of intra-class information and the reliability of class centers. Specifically, a pedestrian attention module is proposed to enable the encoder to pay more attention to the person’s features to relieve interference caused by the shared background style. Furthermore, we propose a priority-distance graph convolutional network (PDGCN) module that employs a graph convolutional network network to predict the priority of a node as a class center and then calculates the distance between nodes with high priority values to screen out the class center nodes. Finally, the encoder features (local) and PDGCN features (context-aware) are combined to perform person Re-ID. The results of experiments on the large-scale public Re-ID datasets verified the effectiveness of the proposed method.

Unmanned aerial vehicle (UAV)-assisted mobile edge computing (MEC) is a promising technology to provide computational services to ground terminal devices (TDs) in remote areas or for emergency incidents with its flexibility and mobility. This paper aims to maximize the UAV’s energy efficiency while considering the fairness of offloading. We formulate this optimization problem by jointly considering the UAV flight time, the UAV 3D trajectory, the TD binary offloading decisions, and the time allocated to TDs, which is a mixed-integer nonlinear programming problem. The problem is transformed into two sub-problems and we propose a PDDQNLP (parametrized dueling deep Q-network and linear programming) algorithm based on the combination of deep reinforcement learning (DRL) and linear programming (LP) to address them. For the first sub-problem, a DRL-based algorithm is used to optimize the TD offloading decisions, the UAV trajectory, and the UAV flight time. The action space is hybrid that contains discrete actions (e.g., binary offloading) and continuous actions (e.g., UAV flight time). Therefore, we parameterize the action space and propose the PDDQN algorithm, which combines the DDPG algorithm for handling the continuous action space and the Dueling DQN algorithm for handling the discrete action space. According to the solution obtained from the first sub-problem, the LP is used to adjust the time allocated to TDs in the second sub-problem. The numerical results show that the PDDQNLP algorithm outperforms its counterparts.

Deep learning methods have shown significant performance in medical image analysis tasks. However, they generally act like ”black box” without explanations in both feature extraction and decision processes, leading to lack of clinical insights and high risk assessments. To aid deep learning in envisioning diseases with visual clues, we propose Representation Group-Disentangling Network (RGD-Net), which can completely disentangle feature space of input X-ray images into several independent feature groups, each corresponding to a specific disease. Taking several semantically related and labeled X-ray images as input, RGD-Net firstly extracts completely group-disentangled representations of diseases through Group-Disentangle Module, which applies group-swap and linking operations to construct latent space by enforcing semantic consistency of attributes. To prevent learning degenerate representations defined as shortcut problem, we further introduce adversarial constricts on mapping from features to diseases, thus avoiding model collapse with former free-form disentanglement. Experiments on chestxray-14 and ChestXpert datasets demonstrate that RGD-Net are effective in predicting diseases with remarkable advantages, which leverage potential factors contributing to different diseases, thus enhancing interpretability in working patterns of deep learning methods.

The autonomous vehicle platoon has the potential to cope with the stress caused by the resource-constrained vehicles‘ demand for processing power and the spread-out deployment of MEC-BS. In this poster, we focus on a multi-platoons scenario for task scheduling. Our objective is to minimize the overall energy consumption subject to the long-term latency constraint. To characterize stochastic properties and deal with coupling between variables, we propose a dynamic task scheduling algorithm based on Lyapunov optimization (LDTS). We theoretically and empirically evaluate the performance of the proposed algorithm, which is illustrated to be significantly better than state-of-the-art and other benchmark approaches in terms of execution latency and energy consumption.

Incremental few-shot semantic segmentation (IFSS) targets at incrementally expanding model's capacity to segment new class of images supervised by only a few samples. However, features learned on old classes could significantly drift, causing catastrophic forgetting. Moreover, few samples for pixel-level segmentation on new classes lead to notorious overfitting issues in each learning session. In this paper, we explicitly represent class-based knowledge for semantic segmentation as a category embedding and a hyper-class embedding, where the former describes exclusive semantical properties, and the latter expresses hyper-class knowledge as class-shared semantic properties. Aiming to solve IFSS problems, we present EHNet, i.e., Embedding adaptive-update and Hyper-class representation Network from two aspects. First, we propose an embedding adaptive-update strategy to avoid feature drift, which maintains old knowledge by hyper-class representation, and adaptively update category embeddings with a class-attention scheme to involve new classes learned in individual sessions. Second, to resist overfitting issues caused by few training samples, a hyper-class embedding is learned by clustering all category embeddings for initialization and aligned with category embedding of the new class for enhancement, where learned knowledge assists to learn new knowledge, thus alleviating performance dependence on training data scale. Significantly, these two designs provide representation capability for classes with sufficient semantics and limited biases, enabling to perform segmentation tasks requiring high semantic dependence. Experiments on PASCAL-5i and COCO datasets show that EHNet achieves new state-of-the-art performance with remarkable advantages.

Abstract The protocol of onion routing constitutes the underpinning of Onion Routing network for anonymous communication. However, since the main idea behind such protocol is to hierarchically peel the IP head and decode each crypted routing address, the topology with regards to path length or long‐term direction cannot be preserved. Owing to topology exposure, lots of attacks such as denial of service and differential flow analysis can be effectively conducted to violate the security of it. Moreover, considering that the path should be constructed by a sender in advance, onion routing is infeasible to be transplanted to ad hoc networks, let alone other defects such as heavy burden about computational complexity and package length. In this paper, a topology‐hiding routing protocol is proposed to address most of the aforementioned problems recurring to homomorphic learning with errors. Security analysis illustrated that nothing but only the IPs of adjacent hops will be revealed to any router. Also the computational complexity on sender side as well as the package length outperforms those of traditional onion routing algorithm in view of comparative simulation.

The Internet of Vehicles (IoV) environment consists of a number of latency-critical and data-intensive application (e.g., real-time video analytics). In this article, we posit the potential of leveraging the sixth-generation (6G) mobile networks to minimize communication delay, particularly for latency-critical task execution. In particular, the 6G-enabled network in boxes (NIBs) deployed in the vehicles can communicate in real time with the edge servers or the NIBs in other vehicles. Although NIBs are capable of providing dynamic and flexible computing resources to support real-time IoV services, there are significant energy costs associated with the communication and computing activities. Seeking to achieve an optimal balance between energy consumption and time cost during service migration, we design a NIB task migration (NTM) method for IoV in this article. In our approach, the IoV framework is designed and the routing mechanism is established. The strength Pareto evolutionary algorithm (SPEA2) is then utilized to determine the migration strategy. Findings from our experiments demonstrate the reliability and efficiency of our proposed approach.

For smart cities, ubiquitous user connectivity and collaborative computation offloading are significant for the ever-increasing information requirements to promote the quality of citizens’ life. In this article, we design an information prefetching architecture, which investigates a hierarchical data storage and selection strategy, including local to edge and edge to cloud. Building on collected data in the social media system or sensor networks, we specifically focus on analyzing mobile terminals’ behaviors to assure the precision of our prefetching strategy in different kinds of information streaming. To assemble edge agents (EAs) prefetching, we also consider the characteristics of wireless backhaul. This scheme is carried out to optimize the EAs prefetching framework by the independent and joint action modules that are based on the theory of deep reinforcement learning (DRL). It paves a better way of collaborative edge computing (CEC) that can be built by using an independent/joint edge-learning model to help and promote the algorithm efficiency and cost-effectiveness. Furthermore, for hiding the information of data transmission between the cloud and the edge servers during data prefetching, this hierarchical scheme is designed as an implicit index maintained by edge servers. Our results show rationales on the obtainable performance of EAs architectures and their reciprocity with the dynamic change of mobile terminals’ requirements.

Motivated by Industry 4.0, 5G-enabled unmanned aerial vehicles (UAVs; also known as drones) are widely applied in various industries. However, the open nature of 5G networks threatens the safe sharing of data. In particular, privacy leakage can lead to serious losses for users. As a new machine learning paradigm, federated learning (FL) avoids privacy leakage by allowing data models to be shared instead of raw data. Unfortunately, the traditional FL framework is strongly dependent on a centralized aggregation server, which will cause the system to crash if the server is compromised. Unauthorized participants may launch poisoning attacks, thereby reducing the usability of models. In addition, communication barriers hinder collaboration among a large number of cross-domain devices for learning. To address the abovementioned issues, a blockchain-empowered decentralized horizontal FL framework is proposed. The authentication of cross-domain UAVs is accomplished through multisignature smart contracts. Global model updates are computed by using these smart contracts instead of a centralized server. Extensive experimental results show that the proposed scheme achieves high efficiency of cross-domain authentication and good accuracy.

The Industrial Internet of Things (IIoT) is the essential component of Industry 4.0. Blockchain is a promising technology for secure data sharing and trustable cooperation between IIoT devices. However, the ever-growing transaction records make it difficult for the storage-limited IIoT devices to join the blockchain network. In this article, an adaptive compression scheme is proposed to decrease the storage volume on each node. In the scheme, the block body is compressed by representing the included transactions as their remainders stored in the distributed nodes. The original transaction could be recovered based on the Chinese remainder theorem. In particular, each node adapts its compression ratio according to its storage resource. The nodes storing more data have advantages in transaction recovery, introducing an incentive mechanism for efficient storage utilization. The theoretical analysis and simulation results show that the proposed scheme can achieve a high compression ratio with good service availability. The proposed scheme dramatically lowers the threshold for IIoT devices to join the blockchain network, which is important for the large-scale application of blockchain in Industry 4.0.

Critical energy infrastructure (CEI) systems are vital to underpin the national economy and social development, but vulnerable to cyber attack and data privacy leakage when distributed machine learning technologies are deployed on them. Although federated learning (FL) has promoted distributed collaborative learning while keeping natural compliance with the privacy protection, it is tremendously difficult to schedule edge nodes of CEI collaboratively when asynchronous FL tasks are applied in CEI system, since the CEI system must make an irrevocable immediate decision on whether to hire a participant who arrives and departs dynamically without knowing future information. In this article, we tackle this issue by designing fairness-aware and time-sensitive task allocation mechanisms in asynchronous FL for CEI. First, we design an optimal multidimensional contract to guarantee the reliability, honesty, and fairness, and maximize the learning accuracy for the fixed deadline scenario. Second, we design a multimetric participant recruitment mechanism to control time consumption for the limited budget scenario, prove that the problem of optimizing this mechanism is NP-hard, and propose an$e$-approximation algorithm accordingly. Finally, extensive experiments using both real-world data and simulated data further demonstrate the effectiveness and efficiency of our proposed mechanisms compared to the state-of-the-art approaches.

The rise of the Internet of Medical Things introduces the healthcare ecosystem in a new digital era with multiple benefits, such as remote medical assistance, realtime monitoring, and pervasive control.However, despite the valuable healthcare services, this progression raises significant cybersecurity and privacy concerns.In this article, we focus our attention on the IEC 60 870-5-104 protocol, which is widely adopted in industrial healthcare systems.First, we investigate and assess the severity of the IEC 60 870-5-104 cyberattacks by providing a quantitative threat model, which relies on Attack Defence Trees and Common Vulnerability Scoring System v3.1.Next, we introduce an intrusion detection and prevention system (IDPS), which is capable of discriminating and mitigating automatically the IEC 60 870-5-104 cyberattacks.The proposed IDPS takes full advantage of the machine learning (ML) and software defined networking (SDN) technologies.ML is used to detect the IEC 60 870-5-104 cyberattacks, utilizing 1) Transmission Control Protocol/Internet Protocol network flow statistics and 2) IEC 60 870-5-104 payload flow statistics.

With the rapid development of wireless communication, traditional cloud computing cannot fully support low-latency services, especially in sensor networks. Mobile edge computing (MEC) can improve the quality of experience of end users and save the energy consumption of mobile end devices by providing computing resources and storage space. However, it may cause discontinuity of services if these mobile end devices roam around different MEC servers’ areas. To solve the aforementioned problem, in this article, we propose an effective edge-intelligent service placement algorithm (EISPA), which transforms the service placement problem into finding a globally optimal solution via nature-inspired particle swarm optimization (PSO). Moreover, we use a shrinkage factor and combine it with the simulated annealing (SA) algorithm to adjust the position of particles in our algorithm, which aims to avoid falling into an optimal local solution to a certain extent. Performance analysis results show that the EISPA is approaching the optimal enumeration collaborative computation offloading algorithm, and system cost under energy constraints is 83.6%, 20.4%, and 20.3% lower than that in Only Local, Finding the Nearest Edge, and the genetic SA-based PSO algorithms, respectively, which proves that the EISPA has better performance.

In ear of smart cities, intelligent medical image recognition technique has become a promising way to solve remote patient diagnosis in IoMT. Although deep learning-based recognition approaches have received great development during the past decade, explainability always acts as a main obstacle to promote recognition approaches to higher levels. Because it is always hard to clearly grasp internal principles of deep learning models. In contrast, the conventional machine learning (CML)-based methods are well explainable, as they give relatively certain meanings to parameters. Motivated by the above view, this paper combines deep learning with the CML, and proposes a hybrid intelligence-driven medical image recognition framework in IoMT. On the one hand, the convolution neural network is utilized to extract deep and abstract features for initial images. On the other hand, the CML-based techniques are employed to reduce dimensions for extracted features and construct a strong classifier that output recognition results. A real dataset about pathologic myopia is selected to establish simulative scenario, in order to assess the proposed recognition framework. Results reveal that the proposal that improves recognition accuracy about two to three percent.

The papers in this special section focus on emerging Internet of Medical Things. Recent advances in advances in healthcare can be experienced with the development of smart sensorial things, Artificial Intelligence (AI), Machine Learning (ML), Deep Learning (DL), edge computing, Edge AI, 6G, cloud computing, and connected healthcare have attracted a great deal of attention and a wide range of views. However, the need to deliver real-time and accurate healthcare services to patients, while reducing costs is a challenging issue [1]. Especially, COVID-19 has recently demonstrated the importance of fast, comprehensive, and accurate intelligent healthcare involving different types of medical, physiological, and epidemiological investigation data to diagnose the virus. Smart health is a real-time, intelligent, ubiquitous healthcare service based on Internet of bioMedical Things (IoMT). With the rapid development of related technologies such as deep learning, edge computing and IoT, smart health is playing vital role in healthcare industry to increase the accuracy, reliability, and productivity of mobile sensory devices.

In the Internet of Vehicles (IoV), the classic TCP/IP still plays an important role for data transmission, traffic control and address assignment. However, with increasing requirements on content retrieve efficiency in IoV, the drawbacks of traditional TCP/IP stacks, such as weak scalability in large networks, low efficiency in dense environment and unreliable addressing in high mobility circumstance, have incurred significant performance degradations in vehicular environments. Fortunately, the emerging Named Data Network (NDN) technology provides a good choice to address above issues in vehicular environment by proving content caching capability with introduced content store module, and boosts the research activity of Vehicular Named Data Network (VNDN) in the last few years. In this paper, to improve the service performance, e.g., reducing the delay of data acquisition, a data caching scheme is proposed by taking the spatial-temporal characteristics of data into account. At first, we divided the data in a VNDN into emergency safety message, traffic efficiency message and service message, according to the application requirements. Then, we analyze the spatial-temporal characteristics of these three message categories and design the caching strategy according to these characteristics. Experimental results from NDNSim platform show that our designed scheme has an approximately 50% performance enhancement compared with Leave Copy Everywhere (LCE), Pro(0.7), and Pro(0.2) data caching protocols in terms of average hit rate, average hop count and average cache replacement times, which verifies the reliability and effectiveness of our proposed data caching scheme.

The sixth-generation network (6G) is expected to achieve a fully connected world, which makes full use of a large amount of sensitive data. Federated Learning (FL) is an emerging distributed computing paradigm. In Vehicular Edge Computing (VEC), FL is used to protect consumer data privacy. However, using FL in VEC will lead to expensive communication overheads, thereby occupying regular communication resources. In the traditional FL, the massive communication rounds before convergence lead to enormous communication costs. Furthermore, in each communication round, many clients upload large quantity model parameters to the parameter server in the uplink communication phase, which increases communication overheads. Moreover, a few straggler links and clients may prolong training time in each round, which will decrease the efficiency of FL and potentially increase the communication costs. In this work, we propose an efficient-communication approach, which consists of three parts, including “Customized”, “Partial”, and “Flexible”, known as FedCPF. FedCPF provides a customized local training strategy for vehicular clients to achieve convergence quickly through a constraint item within fewer communication rounds. Moreover, considering the uplink congestion, we introduce a partial client participation rule to avoid numerous vehicles uploading their updates simultaneously. Besides, regarding the diverse finishing time points of federated training, we present a flexible aggregation policy for valid updates by constraining the upload time. Experimental results show that FedCPF outperforms the traditional FedAVG algorithm in terms of testing accuracy and communication optimization in various FL settings. Compared with the baseline, FedCPF achieves efficient communication with faster convergence speed and improves test accuracy by 6.31% on average. In addition, the average communication optimization rate is improved by 2.15 times.

The integration of advanced sensing, signal processing, deep learning, and edge computing into vehicles is enabling intelligent automated vehicles that can navigate autonomously in various environments. There are several exciting developments in new technologies that may contribute to the improvement of the robustness of autonomous vehicles and thus making them safer on the road. However, the development of suitable context prediction methodologies in order to provide proactive behavior for intelligent transportations remains a challenge. The reason is that future context information, hidden in the raw context traces left by users in the real world, is not immediately accessible to applications. Therefore, sophisticated context prediction approaches are required that could discover and mine patterns (e.g., of a driver’s behavior) from observed context history. The major challenge of a context prediction approach is in the prediction accuracy and prediction expressiveness. Neural networks along with deep-learning methods have shown noticeably better performance in comparison with previous methods regarding the accuracy of the outcomes. However, deep learning also issues more complexity and interpretability problems and, hence, arises serious challenges regarding the verifiability of these approaches. This Special Issue aims to provide the scientific community with a comprehensive overview of innovative technologies, advanced architectures, and potential challenges for context prediction of autonomous vehicles.

Modern transportation is associated with considerable challenges related to safety, mobility, the environment and space limitations. Vehicular networks are widely considered to be a promising approach for improving satisfaction and convenience in transportation. However, with the exploding popularity among vehicle users and the growing diverse demands of different services, ensuring the efficient use of resources and meeting the emerging needs remain challenging. In this paper, we focus on resource allocation in vehicular cloud computing (VCC) and fill the gaps in the previous research by optimizing resource allocation from both the provider’s and users’ perspectives. We model this problem as a multi-objective optimization with constraints that aims to maximize the acceptance rate and minimize the provider’s cloud cost. To solve such an NP-hard problem, we improve the nondominated sorting genetic algorithm II (NSGA-II) by modifying the initial population according to the matching factor, dynamic crossover probability and mutation probability to promote excellent individuals and increase population diversity. The simulation results show that our proposed method achieves enhanced performance compared to the previous methods.

As a key use case of Industry 4.0 and the Smart City, the Internet of Vehicles (IoV) provides an efficient way for city managers to regulate the traffic flow, improve the commuting performance, reduce the transportation facility cost, alleviate the traffic jam, and so on. In fact, the significant development of Internet of Vehicles has boosted the emergence of a variety of Industry 4.0 applications, e.g., smart logistics, intelligent transforation, and autonomous driving. The prerequisite of deploying these applications is the design of efficient data dissemination schemes by which the interactive information could be effectively exchanged. However, in Internet of Vehicles, an efficient data scheme should adapt to the high node movement and frequent network changing. To achieve the objective, the ability to predict short-term traffic is crucial for making optimal policy in advance. In this article, we propose a novel data dissemination scheme by exploring short-term traffic prediction for Industry 4.0 applications enabled in Internet of Vehicles. First, we present a three-tier network architecture with the aim to simply network management and reduce communication overheads. To capture dynamic network changing, a deep learning network is employed by the controller in this architecture to predict short-term traffic with the availability of enormous traffic data. Based on the traffic prediction, each road segment can be assigned a weight through the built two-dimensional delay model, enabling the controller to make routing decisions in advance. With the global weight information, the controller leverages the ant colony optimization algorithm to find the optimal routing path with minimum delay. Extensive simulations are carried out to demonstrate the accuracy of the traffic prediction model and the superiority of the proposed data dissemination scheme for Industry 4.0 applications.

Although mobile crowdsensing (MCS) has become a green paradigm of collecting, analyzing, and exploiting massive amounts of sensory data, existing incentive mechanisms are not effective to stimulate users’s active participation and service contribution in multi-service exchange in MCS due to its specific features: a large number of heterogeneous users have asymmetric service requirements, workers have the freedom to choose sensing tasks as well as participation levels, and multiple sensing tasks have heterogeneous values which may be untruthful declared by the corresponding requesters. To address this issue, this article develops a green Stackelberg-game incentive mechanism to achieve selective fairness, truthfulness, and bounded efficiency while reducing the burden on the platform. First, we model the multi-service exchange problem as a Stackelberg multi-service exchange game consisting of multi-leader and multi-follower, in which each requester as a leader first chooses the reward declaration strategy and thus the payment for each sensing task, each worker as a follower then chooses the sensing plan strategy to maximize her own utility. We next introduce the concept of virtual currency to maintain the selective fairness to balance service request and service provision between users, in which a user earns/consumes virtual currency for providing/receiving services, and thus no one can always get services without providing services. Then, we present two novel algorithms to compute the unique Nash equilibrium for the sensing plan determination game and the reward declaration determination game, respectively, which together forms a unique Stackelberg equilibrium for the proposed game. Afterwards, we theoretically prove that the proposed green Stackelberg-game incentive mechanism achieves the desirable properties of selective fairness, truthfulness, bounded efficiency. Finally, extensive evaluation results are provided to support the validity and effectiveness of our mechanism compared with both baseline and theoretical optimal approaches.

Atrial fibrillation detection (AFD) has attracted much attention in the field of embedded systems. In this study, we propose a sorting fuzzy min-max (SFMM) model, and then develop an SFMM-based embedded system for AF detection. The proposed SFMM model is essentially enhanced the fuzzy min-max (FMM) model that have been successfully applied in many classification fields. In comparison with the typical FMM model, the proposed SFMM model can overcome the limitation of the input order problem encountered in the typical FMM model. The embedded system consists of a control chip and an analog-digital conversion (ADC) chip. The STM32F407 chip is used as the control chip and the ADS1292 chip, which has a high common-mode rejection ratio (CMRR), is used as the ADC chip. A series of machine learning benchmarks are included to evaluate the performance of the SFMM model. Experimental results on AF data further demonstrate the effectiveness of the SFMM-based embedded system.

The advent of Internet of Things (IoT) has bring a new era in communication technology by expanding the current inter-networking services and enabling the machine-to-machine communication. IoT massive deployments will create the problem of optimal power allocation. The objective of the optimization problem is to obtain a feasible solution that minimizes the total power consumption of the WSN, when the error probability at the fusion center meets certain criteria. This work studies the optimization of a wireless sensor network (WNS) at higher dimensions by focusing to the power allocation of decentralized detection. More specifically, we apply and compare four algorithms designed to tackle Large scale global optimization (LSGO) problems. These are the memetic linear population size reduction and semi-parameter adaptation (MLSHADE-SPA), the contribution-based cooperative coevolution recursive differential grouping (CBCC-RDG3), the differential grouping with spectral clustering-differential evolution cooperative coevolution (DGSC-DECC), and the enhanced adaptive differential evolution (EADE). To the best of the authors knowledge, this is the first time that LSGO algorithms are applied to the optimal power allocation problem in IoT networks. We evaluate the algorithms performance in several different cases by applying them in cases with 300, 600 and 800 dimensions.

Although deep learning has achieved great success in object detection recently, scene text detection is still a challenging task, due to inherent difficulties of locating texts in complex scenes. Many approaches adopt inspirations from segmentation to detect arbitrary shaped scene text. However, most segmentation based methods have high computation cost and generally needs a lot of refinements to get accurate results. To ease this problem, we propose a novel single-stage method, i.e., PolarText network, which detects text regions by generating contour points in polar coordinates. PolarText not only relieves the burden of high computation cost by directly regressing contour points instead of pixels, but also fits with intrinsic characteristics of text instances by centers and contours, thus suppressing mislabeling boundary pixels caused by pixel-level labeling. To cope with polar representation, PolarText utilizes Polar IoU loss and polar centerness to generalize effective paradigms from box representation for polar representation. In addition, we add a dedicated bounding box branch to work with text detection since most text instances are approximately rectangular in shape. Compared with the existing methods, the proposed method achieves superior results in both accuracy and efficiency by testing on CTW 1500 and ICDAR 2015 datasets.

In order to predict the unknown image categories, few-shot image classification has recently become a very hot field. However, many methods need a large number of samples to support in order to achieve enough functions. This makes the whole network de amplification to meet a large number of effective feature extraction, and reduces the efficiency of few-shot classification to a certain extent. To solve these problems, we propose a dilate convolutional network with data enhancement. This network can not only meet the necessary features of image classification without increasing the number of samples, but also has a structure that utilizes a large number of effective features without sacrificing efficiency. The cutout structure can enhance the data by adding a fixed area 0 mask matrix in the process of image input. The structure of FAU uses dilate convolution and uses the characteristics of a sequence to improve the efficiency of the network.

n order to predict the unknown image categories, few-shot image classification has recently become a very hot field. However, many methods need a large number of samples to support in order to achieve enough functions. This makes the whole network de amplification to meet a large number of effective feature extraction, and reduces the efficiency of few-shot classification to a certain extent. To solve these problems, we propose a dilate convolutional network with data enhancement. This network can not only meet the necessary features of image classification without increasing the number of samples, but also has a structure that utilizes a large number of effective features without sacrificing efficiency. The cutout structure can enhance the data by adding a fixed area 0 mask matrix in the process of image input. The structure of FAU uses dilate convolution and uses the characteristics of a sequence to improve the efficiency of the network.

Distribution estimation of physical symptom is a powerful statistical technique to monitor and predict the situation of infectious diseases. With the widespread popularization of smart devices, edge computing is an efficient computing paradigm, which takes advantage of these devices to implement the safe and fast collection and distribution analysis of physical symptoms. However, the sharing of individuals' data, especially of sensitive information (e.g., age, symptom), inevitably raises the great privacy concerns of people due to the possible leakage and the illegal utilization. It is necessary and important to not only implement the highly intelligent epidemic analysis, but also guarantee the privacy of data. In this paper, we focus on locally private distribution estimation of physical symptom with attribute constraints in edge computing. At first, we present an edge-computing-based framework for distribution estimation of physical symptoms, in which each component undertakes the respective task. Then, we propose a naive privacy algorithm, i.e., NAIVEPRIVDISTEST, to directly perturb the values of quasi-identifying attributes and physical symptoms. Finally, extensive experiments demonstrate that NAIVEPRIVDISTEST has the accurate distribution estimation under the privacy constraints.

Internet of Things (IoT) is expected to vastly increase the number of connected devices. As a result, a multitude of IoT devices transmit various information through wireless communication technology, such as the Wi-Fi technology, cellular mobile communication technology, low-power wide-area network (LPWAN) technology. However, even the latest Wi-Fi technology is still ready to accommodate these large amounts of data. Accurately setting the contention window (CW) value significantly affects the efficiency of the Wi-Fi network. Unfortunately, the standard collision resolution used by IEEE 802.11ax networks is nonscalable; thus, it cannot maintain stable throughput for an increasing number of stations, even when Wi-Fi 6 has been designed to improve performance in dense scenarios. To this end, we propose a CW control strategy for Wi-Fi 6 systems. This strategy leverages deep learning to search for optimal configuration of CW under different network conditions. Our deep neural network is trained by data generated from a Wi-Fi 6 simulation system with some varying key parameters, e.g., the number of nodes, short interframe space (SIFS), distributed interframe space (DIFS), and data transmission rate. Numerical results demonstrated that our deep learning scheme could always find the optimal CW adjustment multiple by adaptively perceiving the channel competition status. The finalized performance of our model has been significantly improved in terms of system throughput, average transmission delay, and packet retransmission rate. This makes Wi-Fi 6 better adapted to the access of a large number of IoT devices.

The number of smart devices newly connected to the Internet has grown exponentially in recent years. These smart devices are interwoven into huge Internet of Things. There is a contradiction between mass data transmission and communication bandwidth, the distance between supercomputing power and processing object, and the demand of frequent interaction and real-time response. As a new computing paradigm, edge computing processes tasks on computing resources close to data sources. Considering the limited energy of the mobile terminal and the user's demand for low delay, making decisions about tasks executed locally and offloaded to edge computing servers. In the edge environment, resources are dynamically allocated to users on demand, and users need to pay for the resources they actually consume. By considering energy consumption, delay, and price, a user-centered joint optimization loading scheme is proposed to minimize the weighted cost of time delay, energy consumption, and price under the constraint of satisfying the advanced personalized needs of users. The optimization problem is modeled as a mixed-integer nonlinear programming problem, and a branch-and-bound algorithm based on linear relaxation improvement is proposed to solve the problem. Considering the complexity of the algorithm, a particle swarm optimization algorithm based on 0-1 and weight improvement is proposed to solve the problem. Simulation results show that the method proposed in this article can achieve higher performance in terms of delay, energy consumption, and price and provide personalized service for users.

Smart contracts, one of the success stories in blockchain 2.0, have been widely utilized in a broad range of applications, including those involving Internet of Things (IoT). Given the fast-pace nature of the topic, it can be challenging for the research community to keep track of the latest advances. Hence, in this article, we perform a comprehensive, in-depth review of known security challenges (e.g., inherently vulnerable particularities, programming vulnerabilities, and attacks) and potential research opportunities associated with the deploying of smart contracts in an IoT setting. We hope this survey will serve as a starting point for the readers seeking to understand and explore the potential applications of smart contracts.

Although crowdsourcing has emerged as a paradigm for leveraging human intelligence and activity to solve a wide range of tasks, strategic workers will find enticement in their self-interest to free-ride and attack in a crowdsourcing contest dilemma game. Existing incentive mechanisms are not effective to avoid socially undesirable equilibrium due to the following features of competitive crowdsourcing: in the presence of imperfect monitoring, heterogeneous workers with competing interest tend to beat their opponents for larger self-profit, and the fact that they can freely and frequently change their opponents makes the situation much more complicated. Taking these features into consideration, this article proposes a mechanism design problem to enforce cooperation and extort selfish works simultaneously, with the objective of maximizing the requester's utility. To solve the problem, we integrate binary ratings with differential pricing to develop a novel rating protocol. By establishing a mathematical model for the problem and quantifying necessary and sufficient conditions for a sustainable social norm, we provide design guidelines for optimal rating protocols and design a low-complexity algorithm to select optimal design parameters. Finally, extensive evaluation results demonstrate the performance of our proposed rating protocol and reveal how intrinsic parameters impact on design parameters.

As one of the most successful recommendation techniques, neighborhood-based collaborative filtering (CF), which recommends appropriate items to a target user by identifying similar users or similar items, has been widely applied to various recommender systems. Although many neighbor-based CF methods have been put forward, there are still some open issues that have remained unsolved. First, the ever-increasing volume of user–item rating data decreases the recommendation efficiency significantly as a recommender system needs to analyze all the rating data when searching for similar neighbors or similar items. In this situation, users’ requirements on quick response may not be met. Second, in neighbor-based CF methods, more attention is paid to the recommendation accuracy while other key indicators of recommendation performances are often ignored, i.e., recommendation diversity (RD), which probably produces similar or redundant items in the recommended list and decreases users’ satisfaction. Considering these issues, a diversified and scalable recommendation method (called DR_LT) based on locality-sensitive hashing and cover tree is proposed in this article, where the item topic information is used to optimize the final recommended list. We show the effectiveness of our proposed method through a set of experiments on MovieLens data set that clearly shows the feasibility of our proposal in terms of item recommendation accuracy, diversity, and scalability.

Nowadays, large number of cloud-based techniques have been used in industrial control systems (ICS), which also brings many security threats. The emergence of security-aware industrial control has paved the way of security-aware scheduling in cloud-based industrial applications. Actually, most cloud-based industrial applications are time sensitive, which need real-time processing. Edge cloud computing paradigm extends the computing ability of traditional cloud model with low-latency local resources. Thus, heterogeneous clouds that consist of both centralized resources and edge resources may be a promising resource model to provide both scalable and low-latency resources for cloud-based industrial applications. In view of these challenges, in this article, we propose a security-aware dynamic scheduling method for real-time resource allocation in ICS. First, a three-level security model is designed for both tasks and cloud resources in ICS, and a two-tier heterogeneous cloud architecture is introduced. Accordingly, a security-aware scheduling method based on distributed particle swarm optimization is presented for resource allocation with security concerns. To deal with the dynamics of edge resources and the mobility of mobile industrial applications, a dynamic scheduling mechanism based on dynamic workflow model is proposed for real-time optimization. Experimental results validate that the scheduling control policy proposed in this article can achieve a good balance between scheduling performance and security performance.

The automotive industry, a key part of industrial Internet of Things, is now converging with cognitive computing (CC) and leading to industrial cognitive Internet of Vehicles (CIoV). As the major data source of industrial CIoV, social media has a significant impact on the quality of service (QoS) of the automotive industry. To provide vehicular social media services with low latency and high reliability, edge computing is adopted to complement cloud computing by offloading CC tasks to the edge of the network. Generally, task offloading is implemented based on the premise that edge servers (ESs) are appropriately quantified and located. However, the quantification of ESs is often offered according to empirical knowledge, lacking analysis on real condition of intelligent transportation system (ITS). To address the abovementioned problem, a collaborative method for the quantification and placement of ESs, named CQP, is developed for social media services in industrial CIoV. Technically, CQP begins with a population initializing strategy by Canopy and K-medoids clustering to estimate the approximate ES quantity. Then, nondominated sorting genetic algorithm III is adopted to achieve solutions with higher QoS. Finally, CQP is evaluated with a real-world ITS social media data set from China.

An intelligent transportation system (ITS) plays an important role in public transport management, security and other issues. Traffic flow detection is an important part of the ITS. Based on the real-time acquisition of urban road traffic flow information, an ITS provides intelligent guidance for relieving traffic jams and reducing environmental pollution. The traffic flow detection in an ITS usually adopts the cloud computing mode. The edge of the network will transmit all the captured video to the cloud computing center. However, the increasing traffic monitoring has brought great challenges to the storage, communication and processing of traditional transportation systems based on cloud computing. To address this issue, a traffic flow detection scheme based on deep learning on the edge node is proposed in this article. First, we propose a vehicle detection algorithm based on the YOLOv3 (You Only Look Once) model trained with a great volume of traffic data. We pruned the model to ensure its efficiency on the edge equipment. After that, the DeepSORT (Deep Simple Online and Realtime Tracking) algorithm is optimized by retraining the feature extractor for multiobject vehicle tracking. Then, we propose a real-time vehicle tracking counter for vehicles that combines the vehicle detection and vehicle tracking algorithms to realize the detection of traffic flow. Finally, the vehicle detection network and multiple-object tracking network are migrated and deployed on the edge device Jetson TX2 platform, and we verify the correctness and efficiency of our framework. The test results indicate that our model can efficiently detect the traffic flow with an average processing speed of 37.9 FPS (frames per second) and an average accuracy of 92.0% on the edge device.

In Internet of Vehicles (IoV), accurate traffic flow prediction is helpful for analyzing road condition and then timely feedback traffic information to managers as well as travelers. Traditional traffic flow predictions are generally suffering from the performance degradation by over-fitting and manual intervening, which cannot support large-scale and high-dimensional urban road network data. To address this issue, in this paper, a traffic flow prediction framework for urban road network based on deep learning is proposed. Firstly, the feature engineering is introduced to extract the features from a large volume of traffic dataset, with the anomaly nodes eliminated. Next, the big traffic dataset is compressed through the spectral clustering compression scheme. Finally, we designed a hybrid traffic flow prediction scheme based on LSTM (Long Short Term Memory) and Sparse Auto-Encoder (SAE). Experimental results show that our proposed model is superior to other models with an average prediction accuracy approaching 97.7%.

The demand for real-time communication and high performance of the Internet of Vehicles (IoV) system has caused researches to investigate new techniques in edge computing (EC). With the rapid development of the fifth- generation (5G) network, the features of low delay, high reliability and superior communication efficiency can bring historic opportunities for the development of the EC-IoV system. In the 5G-enabled EC-IoV system, extreme densification of 5G base stations (gNBs) provides rapid and reliable network access and information interaction. However, this densification also brings more complex connectivity to the network, which increases the difficulty of resource migration and scheduling for the edge devices. Thus, it is still a challenge to manage the resources of the edge devices under the premise of reducing the energy and time cost in the system while avoiding the situation of overload or underload to maintain the stability of the system. In this article, a 5G-enabled EC-IoV system framework is proposed to enhance the performance of *the existing EC-IoV system. Specific computation offloading in 5G-enabled EC-IoV system is presented under three different cases. Through the above cases, two communication modes are concluded and the corresponding resource allocation strategy is given in this article. The performance of the proposed system is evaluated and compared with the existing system. Finally, future research directions in this area are considered.

Intelligent transportation systems pervasively deploy thousands of video cameras. Analyzing live video streams from these cameras is of significant importance to public safety. As streaming video is increasing, it becomes infeasible to have human operators sitting in front of hundreds of screens to catch suspicious activities or detect objects of interests in real-time. Actually, with millions of traffic surveillance cameras installed, video retrieval is more vital than ever. To that end, this article proposes a long video event retrieval algorithm based on superframe segmentation. By detecting the motion amplitude of the long video, a large number of redundant frames can be effectively removed from the long video, thereby reducing the number of frames that need to be calculated subsequently. Then, by using a superframe segmentation algorithm based on feature fusion, the remaining long video is divided into several Segments of Interest (SOIs) which include the video events. Finally, the trained semantic model is used to match the answer generated by the text question, and the result with the highest matching value is considered as the video segment corresponding to the question. Experimental results demonstrate that our proposed long video event retrieval and description method which significantly improves the efficiency and accuracy of semantic description, and significantly reduces the retrieval time.

The advances of Internet technology has resulted in the rapid and pervasive development of e-commerce, which has not only changed the production and operation mode of many enterprises, but also affected the economic development mode of the whole society. This trend has incurred a strong need to store and process large amounts of sensitive data. The traditional data storage and search solutions cannot meet such requirements. To tackle this problem, in this article, we proposed Consortium Blockchain-based Distributed Secure Search (CBDSS) Scheme over encrypted data in e-Commerce environment. By integrating the blockchain and searchable encryption model, sensitive data can be effectively protected. The consortium blockchain can ensure that only authorized nodes can join the system. To fairly assign nodes for the search tasks, we developed an endorsement strategy in which two agent roles are set up to divide and match the search tasks with the virtual resources according to the load capacity of each node. The security analysis and experiments are conducted to evaluate the performance of our proposed scheme. The evaluation results have proved the reliability and security of our scheme over existing methods.

introduction Share on Introduction to the Special Issue on Fine-grained Visual Computing Editors: Shaohua Wan Zhongnan University of Economics and Law Zhongnan University of Economics and LawView Profile , Zan Gao Qilu University of Technology Qilu University of TechnologyView Profile , Hanwang Zhang Nanyang Technological University Nanyang Technological UniversityView Profile , Xiaojun Chang Monash University Monash UniversityView Profile Authors Info & Claims ACM Transactions on Multimedia Computing, Communications, and ApplicationsVolume 17Issue 1sJanuary 2021 Article No.: 11pp 1–3https://doi.org/10.1145/3447532Online:31 March 2021Publication History 0citation70DownloadsMetricsTotal Citations0Total Downloads70Last 12 Months24Last 6 weeks1 Get Citation AlertsNew Citation Alert added!This alert has been successfully added and will be sent to:You will be notified whenever a record that you have chosen has been cited.To manage your alert preferences, click on the button below.Manage my Alerts New Citation Alert!Please log in to your account Save to BinderSave to BinderCreate a New BinderNameCancelCreateExport CitationPublisher SiteGet Access

Currently, edge computing (EC), emerging as a burgeoning paradigm, is powerful in handling real-time resource provision for Internet of Things (IoT) applications. However, due to the spatial distribution of geographically sparse IoT devices and the resource limitations of EC units (ECUs), the resource utilization of corresponding edge servers is relatively insufficient and the execution performance is ineffective to some extent. A privacy leakage, including personal information, location, media data, etc., during the transmission process from IoT devices to edge servers severely restricts the application of ECUs in IoT. To address these challenges, a two-phase offloading optimization strategy is put forward for joint optimization of offloading utility and privacy in EC enabled IoT. Technically, a utility-aware task offloading method, named UTO, is devised first to obtain the goal of maximizing the resource utilization of ECUs and minimizing the implementation time cost. Then a joint optimization method, named JOM, for utility and privacy tradeoffs is designed to balance the privacy preservation and execution performance. Eventually, the experimental evaluations are designed to illustrate the efficiency and reliability of UTO and JOM.

Cloud computing is a formidable paradigm to provide resources for handling the services from Industrial Internet of Things (IIoT), such as meteorological industry. Generally, the meteorological services, with complex interdependent logics, are modeled as workflows. When any of the computing nodes for hosting the meteorological workflows fail, all sorts of consequences (e.g., data loss, makespan enlargement, performance degradation, etc.) could arise. Thus recovering the failed tasks as well as optimizing the makespan and the load balance of the computing nodes is still a critical challenge. To address this challenge, a dynamic resource provisioning method (DRPM) with fault tolerance for the data-intensive meteorological workflows is proposed in this article. Technically, the Virtual Layer 2 (VL2) network topology is exploited to build meteorological cloud infrastructure. Then, the nondominated sorting genetic algorithm II (NSGA-II) is employed to minimize the makespan and improve the load balance. Finally, comprehensive experimental analysis of DRPM are proceeded.

In this paper, we propose a fully automatic image colorization method for grayscale images using neural network and optimization. For a determined training set including the gray images and its corresponding color images, our method segments grayscale images into superpixels and then extracts features of particular points of interest in each superpixel. The obtained features and their RGB values are given as input for, the training colorization neural network of each pixel. To achieve a better image colorization effect in shorter running time, our method further propagates the resulting color points to neighboring pixels for improved colorization results. In the propagation of color, we present a cost function to formalize the premise that neighboring pixels should have the maximum positive similarity of intensities and colors; we then propose our solution to solving the optimization problem. At last, a guided image filter is employed to refine the colorized image. Experiments on a wide variety of images show that the proposed algorithms can achieve superior performance over the state-of-the-art algorithms.

In recent years, view-based 3D model retrieval has become one of the research focuses in the field of computer vision and machine learning. In fact, the 3D model retrieval algorithm consists of feature extraction and similarity measurement, and the robust features play a decisive role in the similarity measurement. Although deep learning has achieved comprehensive success in the field of computer vision, deep learning features are used for 3D model retrieval only in a small number of works. To the best of our knowledge, there is no benchmark to evaluate these deep learning features. To tackle this problem, in this work we systematically evaluate the performance of deep learning features in view-based 3D model retrieval on four popular datasets (ETH, NTU60, PSB, and MVRED) by different kinds of similarity measure methods. In detail, the performance of hand-crafted features and deep learning features are compared, and then the robustness of deep learning features is assessed. Finally, the difference between single-view deep learning features and multi-view deep learning features is also evaluated. By quantitatively analyzing the performances on different datasets, it is clear that these deep learning features can consistently outperform all of the hand-crafted features, and they are also more robust than the hand-crafted features when different degrees of noise are added into the image. The exploration of latent relationships among different views in multi-view deep learning network architectures shows that the performance of multi-view deep learning outperforms that of single-view deep learning features with low computational complexity.

Abstract In order to increase the automatic quality control level in the textile industry, depending on the big data collected by the Internet of things of the textile factories, this paper proposes a novel visual saliency–based defect detection algorithm, which has the capability of automatically detecting defect in both nonpatterned and patterned fabrics. The algorithm employs the histogram features extracted from the saliency maps to detect the fabric defects. The algorithm involves three main steps: (1) saliency map generation to highlight the defective regions and suppress the defect‐free regions, (2) saliency histogram features extraction and selection to obtain the feature vectors that can effectively discriminate between the defective and defect‐free fabric images, and (3) fabric defect detection using a two‐class support vector machine classifier that has been trained using sets of feature vectors extracted from defective and defect‐free fabric samples. Experimental results show that our method yields accurate detections, outperforming other state‐of‐the‐art algorithms.

Seeking to address the challenges associated with high-dimensional complex time series representations of recurrent neural networks, such as low generalization ability and long training time, a hybrid neural network based on a deep belief network (DBN) is proposed in this paper to facilitate time series predictions for the Internet of Things. In our approach, we integrate both a DBN and a recurrent neural network with the gated recurrent unit as the activation unit. First, we implement unsupervised pretraining through the DBN and then supervise the curve fitting using the recurrent neural network. Finally, the hybrid neural network is learned and can make predictions. The experimental results show that the hybrid neural network has a stronger historical learning ability than two other widely used recurrent neural networks and can effectively reduce the number of iterations required by the recurrent neural network, thereby reducing the overall learning time.

Human actions are often captured by multiple cameras (or sensors) to overcome the significant variations in viewpoints, background clutter, object speed, and motion patterns in video surveillance, and action recognition systems often benefit from fusing multiple types of cameras (sensors). Therefore, adaptive fusion of the information from multiple domains is mandatory for multiview human action recognition. Two widely applied fusion schemes are feature-level fusion and score-level fusion. We point out that limitations still exist and there is tremendous room for improvement, including the separate computation of feature fusion and action recognition, or the fixed weights for each action and each camera. However, previous fusion methods cannot accomplish them. In this paper, inspired by nature, the above limitations are addressed for multiview action recognition by developing a novel adaptive fusion and category-level dictionary learning model (abbreviated to AFCDL). It can jointly learn the adaptive weight for each camera and optimize the reconstruction of samples toward the action recognition task. To induce the dictionary learning and the reconstruction of query set (or test samples), the induced set for each category is built, and the corresponding induced regularization term is designed for the objective function. Extensive experiments on four public multiview action benchmarks show that AFCDL can significantly outperforms the state-of-the-art methods with 3% to 10% improvement in recognition accuracy.

The Internet of Vehicles (IoV) is composed of road side units (RSUs) nodes and vehicle nodes. Because the limited bandwidth of RSUs can not satisfy massive requests from vehicle nodes, offloading technology is proposed. Generally, spare vehicle nodes (SVs) with redundant resources are motivated to cache data from RSUs. However, due to the selfishness, the participation of SVs is generally low. Hence, an effective incentive mechanism to motivate SVs to offload is important. There are two problems with current incentive mechanisms: 1) all of them ignore loss aversion, which is the act of preferring avoidance of loss over acquiring equivalent gains. This act can lead to deviation in the final decision and 2) no attention is paid to the critical effects that initial allocation of property have on the final resource allocation. However, researches on behavior economics have found that loss aversion exists and the final resource allocation is affected by the initial configuration. Therefore, we propose an incentive mechanism called reverse auction based on endowment effect (RABEE). To the first problem, we introduce the endowment effect from behavior economics, and propose a term called endowment compensation for increasing the participation rate of SVs. For the second problem, by changing the initial allocation of endowment compensation, we illustrate through theoretical analysis and simulations the significance the initial configuration has in terms of the final resource allocation. Simulations show that RABEE increases the average utility of SVs by 6.8% compared with the conventional scheme, and it also improves participation rate by 2%.

In order to more effectively eliminate the disturbance of vibration signal to ensure the security monitoring of stacker be more accurate in Industrial Internet of Things (IIoT), a cooperative denoising algorithm with interactive dynamic adjustment function was constructed and proposed. First, some basic theories such as EMD, EEMD, LMS, and VSLMS were introduced in detail according the characteristics of stacker in IIoT. Meanwhile, the advantages and disadvantages of varieties of algorithms have been analyzed. Secondly, based on the traditional VSLMS-EEMD, an improved VSLMS-EEMD was proposed. Thirdly, to guarantee the denoising effect of security monitoring in IIoT, a cooperative denosing model and framework named as IDVSLMS-EEMD was designed and constructed based on the advantages of LMS, VSLMS, and improved VSLMS-EEMD. In addition, the assignment rules and models of the corresponding weight coefficients were also set up according to the features of the error signal of denoising process in IIoT. At the same time, we have designed a cooperative denoising algorithm with interactive dynamic adjustment function. And some evaluated indexes such as NSR and SDR were selected and introduced to evaluate the effectiveness of the different algorithms. Thirdly, some simulation examples and real experiment examples of stacker running signals under abnormal condition, which has been developed and applied in Power Grid of China, was used to verify and simulate the effectiveness of our presented algorithm. The experiment comparison results have shown that our algorithm can improve the denosing effect. Finally, some conclusions were discussed and the directions for future engineering application were also pointed out.

The increasing number of web APIs registered in service sharing communities (e.g., http://ProgrammableWeb.com that provides a platform benefiting the social interactions between different software developers) has provided a promising way to quickly build various apps with diverse functions. Generally, an app developer can manually discover, select, and compose a set of appropriate web APIs to build a new app satisfying the developer's functional and nonfunctional business requirements, economically and conveniently. However, the above manual web APIs selection process is usually time-consuming and cumbersome as most app developers often do not have much background knowledge of candidate web APIs. Moreover, the manually selected web APIs cannot always guarantee to be integrated successfully as the compatibilities between different web APIs are often varied and not validated. In view of these challenges, we define a weighted APIs correlation graph (W-ACG) in this paper to model the APIs functions and compatibilities. Furthermore, we propose a novel web APIs recommendation approach named Keywords-based and Compatibility-aware APIs Recommendation (K-CAR) based on the defined W-ACG. Through analyzing the input keywords describing the functions expected by an app developer, K-CAR can return the app developer a set of optimal APIs that are not only functional-qualified but also compatibility-guaranteed. Extensive experiments are deployed on 18 478 real-world web APIs and 6146 real-world apps to evaluate the usefulness and efficiency of K-CAR.

Accurate and automatic organ segmentation is critical for computer-aided analysis towards clinical decision support and treatment planning. State-of-the-art approaches have achieved remarkable segmentation accuracy on large organs, such as the liver and kidneys. However, most of these methods do not perform well on small organs, such as the pancreas, gallbladder, and adrenal glands, especially when lacking sufficient training data. This paper presents an automatic approach for small organ segmentation with limited training data using two cascaded steps-localization and segmentation. The localization stage involves the extraction of the region of interest after the registration of images to a common template and during the segmentation stage, a voxel-wise label map of the extracted region of interest is obtained and then transformed back to the original space. In the localization step, we propose to utilize a graph-based groupwise image registration method to build the template for registration so as to minimize the potential bias and avoid getting a fuzzy template. More importantly, a novel knowledge-aided convolutional neural network is proposed to improve segmentation accuracy in the second stage. This proposed network is flexible and can combine the effort of both deep learning and traditional methods, consequently achieving better segmentation relative to either of individual methods. The ISBI 2015 VISCERAL challenge dataset is used to evaluate the presented approach. Experimental results demonstrate that the proposed method outperforms cutting-edge deep learning approaches, traditional forest-based approaches, and multi-atlas approaches in the segmentation of small organs.

Due to the problem of attribute redundancy in meteorological data from the Industrial Internet of Things (IIoT) and the slow efficiency of existing attribute reduction algorithms, attribute reduction based on a genetic algorithm for the coevolution of meteorological data was proposed. The evolutionary population was divided into two subpopulations: one subpopulation used elite individuals to assist crossover operations to increase the convergence speed of the algorithm, and the other subpopulation balanced the population diversity in the evolutionary process by introducing a random population; these two subpopulations completed the evolutionary operations together. With the TSDPSO-AR algorithm and ARAGA algorithm, the attribute reduction operation for precipitation in meteorological data was performed. The results showed that the proposed algorithm maintained the diversity of the population during evolution, improved the reduction performance, and simplified the information system.

Because the existing approaches for diagnosing sensor networks lead to low precision and high complexity, a new fault detection mechanism based on support vector regression and neighbor coordination is proposed in this work. According to the redundant information about meteorological elements collected by a multisensor, a fault prediction model is built using a support vector regression algorithm, and it achieves residual sequences. Then, the node status is identified by mutual testing among reliable neighbor nodes. Simulations show that when the sensor fault probability in wireless sensor networks is 40%, the detection accuracy of the proposed algorithm is over 87%, and the false alarm ratio is below 7%. The detection accuracy is increased by up to 13%, in contrast to other algorithms. This algorithm not only reduces the communication to sensor nodes but also has a high detection accuracy and a low false alarm ratio. The proposed algorithm is suitable for fault detection in meteorological sensor networks with low node densities and high failure ratios.

The key for location‐based service popularization in vehicular environment is security and efficiency. However, due to the constrained resources in vehicle‐mounted system and the distributed structure of fog computation, disposing of the conflicts between real‐time implementation and user’s privacy remains an open problem. Aiming at synchronously preserving the position information for users as well as the data proprietorship of service provider, an efficient location‐based querying scheme is proposed in this paper. We argue that a recent scheme proposed by Jannati and Bahrak is time‐consuming and vulnerable against active adaptive corruptions. Thus accordingly, a postquantum secure oblivious transfer protocol is devised based on efficient NTRU cryptosystem, which then serves as the understructure of a complete location‐based querying scheme in ad hoc manner. The security of our scheme is proved under universal composability frame, while performance analysis is also carried out to testify its efficiency.

This paper concerns the design of classifiers for medical decision making, and proposes a novel probabilistic neural network classifier with the assistance of a genetic algorithm. Unlike the conventional probabilistic neural networks that use all patterns in data sets as hidden nodes, the proposed neural network adopts some centers through clustering algorithms and the output of data set as the hidden nodes. Comparing with the conventional probability neural network classifiers, the proposed approach significantly decreases the time consuming. Furthermore, a genetic algorithm is utilized to optimize feature selection from the data set. Experimental results are presented on several benchmarks illustrating the relationship between selected features and disease.

Wireless sensor networks deployment and operation are most likely to take place under hazardous conditions. One extreme scenario is the deployment of a wireless sensor network in a mountainous and forested region in which a fire has ignited, for the purpose of localizing and/or tracking, in real time, its spread. Some methods developed in this paper are expected to provide superior performances under these conditions. Existing fires in the deployment region will also affect the nodes’ coverage because the nodes that will fall in the middle of the fire will likely cease to operate. In our work, we develop a coverage hole bypassing algorithm for storing and maintaining holes-information in the network based on which a boundary node can provide the communication path which is more efficient to try to bypass a hole. Simulation results are presented to illustrate the proposed method and evaluate its run-time and message count per node. We show that these routing can be used in many applications.

Wireless sensor networks deployment and operation are most likely to take place under hazardous conditions. One extreme scenario is the deployment of a wireless sensor network in a mountainous and forested region in which a fire has ignited, for the purpose of localizing and/or tracking, in real time, its spread. Some methods developed in this paper are expected to provide superior performances under these conditions. Existing fires in the deployment region will also affect the nodes' coverage because the nodes that will fall in the middle of the fire will likely cease to operate. In our future work, we show that these routing can be used in many applications.