The widespread integration of Artificial Intelligence (AI) in sixth-generation Internet of Things (6G-IoT) applications, introduces significant challenges for ensuring the trustworthy and dependability of AI models. The "black-box" characteristic of numerous Deep Neural Networks (DNNs) creates a notable obstacle for confirming their safety in intricate, ever-changing environments. Consequently, there is a need for extensive testing, requiring the gathering and labeling of a large number of test cases, a process that is both time-intensive and resource-consuming. While previous studies have adapted neuron coverage criteria for steering test case generation in DNNs. Yet, these criteria are white-box measures requiring access to model states and presenting their practical limitations. Conversely, black-box metrics, which focus on outputs, present a more feasible approach. Among these, black-box diversity metrics evaluate model robustness by generating diverse test cases, eliminating the need for internal model details. This paper presents a test case generation framework centered on diversity, known as BDTest. BDTest enhances test adequacy through five stages: (1) Mapping feature vectors extracted from an initial set of seed images onto a low-dimensional manifold utilizing UMAP; (2) Detecting sparse regions using DBSCAN; (3) Sampling key points from these regions via Latin Hypercube Sampling; (4) Reconstructing latent features and generating new images through ICA and GAN inversion; and (5) Measuring the diversity of the generated set using metrics such as the Log-Determinant. Experiments demonstrate that BDTest significantly improves test set diversity and error detection performance, achieving error rates of 59.36%, 59.76%, and 67.03% on VGG19, DenseNet121, and MobileNetV2, respectively, outperforming DeepXplore by an average of 12.43% and DLFuzz by 9.95% across all tested models. When retrained with the generated test cases, the model demonstrated improved accuracy on the original test set, alongside a significant enhancement in accuracy on the natural adversarial test set.

Along with the development of the Industrial Internet of Things, human activity recognition (HAR) has received widespread attention in many fields. Support Vector Machine (SVM), is widely used by researchers for human activity recognition. However, the inherent difference of signal properties from different sensors and various orientations is potentially lost when using the vector-based SVM for human activity recognition. What's more, the outlier sensitivity problem of the standard SVM reduces the accuracy of human activity recognition. To tackle this problem, we present a tensor-based feature representation model and a weighted support tensor machine (WSTM) for human activity recognition. Specifically, tensor-based representations are first used to model features from different sensors and various orientations to retain the latent relationship. In addition, the weighted support tensor machine is proposed to classify the human activities in tensor space while avoiding the outlier sensitivity problem. Experimental results demonstrate the proposed WSTM algorithm.

The digitalization of traffic networks has spurred the development of intelligent transportation systems. By utilizing reinforcement learning for dynamic traffic optimization, it efficiently handles real-world traffic complexities. However, as the demand for real-time, high-efficiency tasks increases, relying solely on reinforcement learning struggles to meet both goals. Integrating reinforcement learning with mobile communication technology offers a promising solution for efficient, low-overhead traffic networks. As an important physical layer technology for Integrated Sensing and Communications Systems, Spectrally Efficient Frequency Division Multiplexing (SEFDM) addresses the communication overhead challenge in reinforcement learning-enabled optimization. However, the main challenge of SEFDM is eliminating the inter-carrier interference (ICI) caused by non-orthogonal modulation. Considering that existing post-interference cancellation methods fail due to the ill-conditioning of the generalized channel matrix, which cannot be directly inverted, we propose a nonlinear precoding algorithm at the transmitter, instead of post-cancellation, that effectively eliminates interference and improves transmission reliability. We firstly use a nonlinear feedback structure to avoid power boost and error propagation. Besides that, Geometric Mean Decomposition (GMD) based interference matrix decomposition algorithm is used in the proposed precoding scheme to avoid matrix singularity and obtain diversity gain. Finally, the numerical results show that the proposed precoding method can achieve higher order QAM SEFDM signaling with higher spectral efficiency and get comparative BER performance.

Federated learning (FL) enables collaborative training among multiple clients while preserving data privacy. However, its practical application is significantly limited by two major challenges: statistical heterogeneity and data distribution drift. Statistical heterogeneity causes the direction of local model updates to deviate from the global training objective, while data distribution drift leads to a mismatch between local models and their cluster models. To address these challenges, this paper proposes an adaptive clustered federated learning framework, Fed-CM. Initially, by capturing the dynamic patterns of personalized layer parameters in clients' models, Fed-CM effectively characterizes the correlations and distributional similarities among clients, reflecting the underlying statistical heterogeneity. Subsequently, this framework leverages client similarities to construct an undirected graph and adaptively performs effective cluster discovery with minimal dependence on hyperparameters. Furthermore, a monitoring strategy tracks the deviation between clients’ update directions and the dominant update direction of their clusters and then adaptively migrates clients experiencing data drift. Such a dynamic strategy helps maintain intra-cluster homogeneity and addresses the mismatch between local models and their cluster models. Compared to other state-of-the-art methods, experimental results on multiple datasets demonstrate that the proposed Fed-CM framework effectively addresses the challenges posed by statistical heterogeneity and data drift, significantly improving the performance and robustness of federated learning models.

This research introduces the Density-Clustering based Aggregation for Personalized Federated Learning (DCPFL) algorithm, which utilizes DBSCAN clustering to enhance model accuracy in AI-enabled aerial and edge computing contexts, particularly for UAVs. The DCPFL framework promotes model sharing among clients, fostering the development of personalized and optimized models. DBSCAN is beneficial in automatically determining cluster numbers using EPS neighborhoods and MinPts, with parameter optimization achieved through cross-experimental analysis. We further refined the model exchange mechanism by integrating a moving average prediction model to optimize the timing of these exchanges. Tests conducted on three public datasets covering two different machine learning tasks show that DCPFL surpasses existing methods, offering greater accuracy and enhanced adaptability in varied data environments. Implementing this algorithm in UAV networks leverages AI capabilities in aerial and edge computing to efficiently balance personalized modeling requirements with high performance, showcasing its potential to push federated learning forward in complex and dynamic settings.

With the rapid advancement of sixth-generation (6G) network communication technology, improvements in data transmission rates, latency, and reliability have driven substantial growth in Internet of Vehicles (IoV) applications. Among these, the integration of 6G-enabled extremely large-scale antenna arrays (ELAAs) has extended the range of near-field (NF) communication, enabling their application in IoV to facilitate efficient and accurate environmental sensing. Through NF communication, vehicles can achieve high-accuracy localization and perception by analyzing the signal phase, channel state information, and beamforming calculations. However, positioning and sensing tasks place substantial computational and energy demands on edge devices, often exceeding traditional capacity limits. To address this challenge, task offloading has emerged as a solution, with mobile edge computing (MEC) offering a lower-latency alternative to centralized cloud computing by processing tasks at the network edge. Despite these advantages, MEC’s limited resources present challenges as the number of connected vehicles increases. Existing approaches to resource allocation often overlook the varied accuracy requirements of IoV tasks, where high-accuracy tasks like indoor navigation require stringent performance standards, while lower-accuracy tasks may tolerate reduced precision to save resources. Motivated by this, we propose an accuracy-based classification scheme for IoV positioning and sensing tasks, which dynamically adjusts accuracy requirements to reduce delay and energy consumption. Our approach maps total energy, accuracy loss, and delay to an overall quality of service (QoS) metric, and employs an optimization algorithm that leverages gradient descent and greedy strategies to balance resource allocation and accuracy selection. Extensive simulations demonstrate the effectiveness of the proposed scheme in reducing delay and energy consumption while maintaining high accuracy, significantly outperforming benchmark strategies.

Federated learning (FL) is a novel paradigm for distribute edge intelligence for the Internet-of-Vehicles (IoV) application, which can enable superior performance in model training without the need to share local data. However, in the actual architecture of FL, the existence of nonindependent and identically distributed (non-IID) data at the edge device, along with the involvement of randomly participating distributed nodes, can result in model bias and a subsequent decrease in overall performance. To solve this problem, a new federated self-supervised learning method based on prototypes clustering contrastive learning (FedPCC) is proposed, which can effectively addresses the issue of asynchronous edge training and global model bias by introducing an unsupervised prototypes layer. The prototypes layer maps edge features to a global space and performs clustering, facilitating the new aggregation method of global prototypes on the server. Then, models from other components are aggregated based on data weight. Besides that, during the parameter deployment phase, we replace the prototype layer to acquire global knowledge, while employing momentum updates to preserve the local knowledge of the other components. Finally, to assess the efficacy of our proposed approach, we carried out comprehensive experiments across the various data sets. The findings show that our method gains state-of-the-art performance, which also validates its effectiveness.

Spatiotemporal signal forecasting is vital for promoting intelligence management in Internet of Everything applications. Benefiting from the powerful representation ability of deep learning, DNN based methods have shown promising performance in spatiotemporal forecasting. However existing methods perform suboptimally in few-shot distribution shift scenarios. As the representation ability results in stable fitting, the model struggles to adapt to discrepancy in the data domain. And it is challenging to modify the mapping relationship between the data and the latent space of representation with few-shot distributionally shifted target data. In this case, we propose a distribution centric prompt based transfer learning framework, which transforms the data distribution information into model-interpretable prompt embeddings confused with spatiotemporal sematic information in Intermediary Bridging Space which serves as a mediator between the data domain and the latent space. Thus the downstream model can learn a target distribution aligned representation regulated by the mediator. Though experiments on real-world datasets, we verify the effectiveness and extensibility of the proposed method.

Digital twin (DT) is a promising technology for responding to Industry 4.0 and realizing comprehensive automation and virtualization. In the Web3.0-powered 5G/6G era, the expansion of the industrial data and closer interaction among cross-industrial entities pose new security challenges for DT industrial systems. As a prevalent computing paradigm, Graph Anomaly Detection provides an effective solution to ensure the security of DT industrial systems. However, the existing unsupervised graph anomaly detection methods tend to treat multiple graph attributes in isolation during the reconstruction process, resulting in insufficient semantics and suboptimal reconstruction performance. To overcome these challenges, we propose a multiattribute contrastive learning framework, which realizes graph anomaly detection by capturing both graph attribute patterns and their hidden relationships. First, we use an improved multiattribute aligned reconstruction approach to represent the anomaly information effectively. Besides that, the positive instance aggregation-based contrastive constraints are proposed, which can reduce the loss generated by mappings between different data dimension in feature representation space. Finally, to verify our proposal, extensive experiments have been conducted on five benchmark datasets, and the results show that our method obtains the state-of-the-art performance.

Federated learning (FL) offers a promising distributed paradigm for managing massive data from Industry 5.0 devices while preserving privacy. However, significant challenges arise from inherent system heterogeneity and data heterogeneity across devices. These factors severely impede FL performance, leading to slow convergence and potential degradation of the global model’s accuracy. Random client selection strategies are often insufficient to mitigate these issues effectively. To address these limitations, we propose FedHRL: a heuristic-guided pointer network reinforcement learning framework for joint client selection and bandwidth allocation in FL. Specifically, our heuristic-guided soft actor–critic algorithm employs a transformer-based pointer network within its action network to tackle the combinatorial optimization problem of sequentially selecting clients and allocating bandwidth. This network identifies the optimal next client based on prior selections and available bandwidth constraints. Furthermore, to accelerate RL convergence and enhance policy effectiveness, we integrate a particle swarm optimization-based bandwidth reallocation strategy, which refines the RL agent’s bandwidth allocation decisions, feeding the optimization results back as an enhanced reward signal to expedite learning and improve overall performance. Experiments demonstrate that FedHRL accelerates FL training convergence while maintaining high model accuracy in heterogeneous environments.

Interest in Intelligent Transportation Systems (ITS) has increased significantly with the development of 6G. Owning an extremely high transmission speed, 6G is able to support low-latency service for edge-intelligence applications by Machine Learning(ML) techniques. However, traditional centralized learning is not suitable for this scenario due to the requirement for users to upload local data to the server, which can compromise data privacy. To overcome this challenge, Federated Learning (FL) and Split Learning (SL), as progressive distributed learning techniques, have been proposed as a solution. They enable the training of ML models while preserving data privacy. However, conventional FL has poor convergence when data heterogeneity occurs, also fails to meet personalized demands. To address these issues, We propose a novel personalized Federated Learning(pFL) framework, which trains models in SL and collaborates in FL. It offers a personalized solution for each client while retaining a global solution for newcomers. Experimental results demonstrate that our method outperforms other advanced baselines on benchmark datasets.

The upcoming 6G technology, with its high speed and low latency, is poised to become a foundational technology for intelligent transportation systems. To handle the massive data generated by connected vehicles in 6G environments, federated learning methods are essential. However, traditional centralized federated learning approaches still face challenges related to data and device heterogeneity, which significantly affects training efficiency. To address these challenges, we propose FedCPC, a context-based adaptive pruning clustered federated learning method. Based on the positive correlation between similar data distributions and model representations, we use centralized kernel alignment (CKA) to group clients with similar data distributions, thus reducing the impact of data heterogeneity. Furthermore, we introduce a context-aware random forest multi-armed bandit method to determine appropriate pruning rates based on device capabilities and historical performance which addresses device heterogeneity concerns. Experimental results on open-source datasets demonstrate that FedCPC outperforms traditional FL methods in both learning efficiency and communication effectiveness.

Uncrewed Aerial Vehicle (UAV) is extensively employed as a mobile base station in areas with inadequate cellular infrastructure to enhance the freshness of vehicle sensors. The Age of Information (AoI) is a metric utilized to characterize the freshness of information produced by vehicle sensors. This paper investigates the use of Uncrewed Aerial Vehicles (UAVs) as mobile base stations to enhance the freshness of vehicle sensor information in areas with inadequate cellular infrastructure. We focus on minimizing the Age of Information (AoI) and UAV energy consumption in a Vehicle-to-UAV (V2U) network within the Manhattan scenario. The challenge lies in jointly optimizing UAV trajectories and vehicle data packet scheduling amidst high vehicle mobility and limited communication range. To address this issue, we employ Reinforcement Learning (RL) to formulate the problem as a Markov Decision Process (MDP), proposing a Dueling Double Deep Q-Network (D3QN) method for trajectory and scheduling optimization. We also introduce Priority Experience Replay (PER) to improve reward acquisition for the UAV, addressing the issue of sparse rewards due to the expansive space for vehicle movement. Simulation results provide empirical evidence supporting the efficacy of the proposed algorithm in comparison to baseline policies.

Computational offloading in mobile edge computing (MEC) systems provides an efficient solution for resource‐intensive applications on devices. However, the frequent communication between devices and edge servers increases the traffic within the network, thereby hindering significant improvements in latency. Furthermore, the benefits of MEC cannot be fully realized when the communication link utilized for offloading tasks experiences severe attenuation. Fortunately, reconfigurable intelligent surfaces (RISs) can mitigate propagation‐induced impairments by adjusting the phase shifts imposed on the incident signals using their passive reflecting elements. This paper investigates the performance gains achieved by deploying multiple RISs in MEC systems under energy‐constrained conditions to minimize the overall system latency. Considering the high coupling among variables such as the selection of multiple RISs, optimization of their phase shifts, transmit power, and MEC offloading volume, the problem is formulated as a nonconvex problem. We propose two approaches to address this problem. First, we employ an alternating optimization approach based on semidefinite relaxation (AO‐SDR) to decompose the original problem into two subproblems, enabling the alternating optimization of multi‐RIS communication and MEC offloading volume. Second, due to its capability to model and learn the optimal phase adjustment strategies adaptively in dynamic and uncertain environments, deep reinforcement learning (DRL) offers a promising approach to enhance the performance of phase optimization strategies. We leverage DRL to address the joint design of MEC‐offloading volume and multi‐RIS communication. Extensive simulations and numerical analysis results demonstrate that compared to conventional MEC systems without RIS assistance, the multi‐RIS‐assisted schemes based on the AO‐SDR and DRL methods achieve a reduction in latency by 23.5% and 29.6%, respectively.

The emerging field of Artificial Intelligence of Things (AIoT) is propelled by the remarkable success of deep learning and hardware evolution, which has a significant impact on our daily lives. However, because of their notorious computing resource intensity, the widespread deployment of AIoT devices requires substantial electricity consumption as support, inevitably escalating energy consumption, and ultimately leads to a significant carbon emissions. Existing research on neural network compression and acceleration struggles to achieve energy-efficient inference on resource-constrained AIoT devices. To address this issue, we propose a software-hardware co-design approach that integrates advanced neural network optimization techniques with hardware power management capabilities to enable energy-efficient inference and ultimately achieve sustainable AIoT. We introduce a lightweight split and refinement block that adaptively reduces redundant computation in both channel and spatial dimensions. Several early exit (EE) branches are added to the backbone, which are controlled by a policy-based EE predictor. With the predicted EE index, a curve-fitting-based frequency scaling algorithm is presented to calculate the optimal frequency that minimizes energy overhead while maintaining latency constraints. Extensive experiments on CIFAR and CINIC classification tasks validate that our proposed method consistently reduces energy consumption for neural network inference while outperforming other competitive methods.

As the Augmented Intelligence of Things (AIoT) advances within vehicle-road coordination systems, challenges related to road traffic data transmission and processing are being increasingly addressed. However, this progress also brings significant risks of privacy data leakage. Federated learning (FL), a distributed machine learning paradigm, effectively safeguards client data privacy by allowing multiple participants to collaboratively train models while keeping their data localized. Despite its benefits, FL faces challenges, such as model parameter leakage and Byzantine attacks. To tackle these issues, this article introduces an anonymous and efficient FL framework for vehicle-road coordination systems (AEFL), designed to ensure a secure and reliable vehicle data transmission process. This architecture incorporates a novel group pairing onion routing protocol, which leverages pairing cryptography principles for hierarchical data encryption. During the routing process, relay group nodes decrypt the corresponding layer, ensuring both data confidentiality and node anonymity. Additionally, a sampling method is proposed to accurately identify Byzantine vehicle nodes, enhancing the precision of FL aggregation without compromising overall model performance. Experimental results show that AEFL outperforms the classic TOR anonymous routing protocol, achieving a 100% message delivery rate more quickly. Under the same conditions, the anonymity of the source node and the destination node improves by 3.9% and 1.9%, respectively. When half of the nodes are compromised, path anonymity can be increased by 24.8%. Furthermore, our framework excels in FL aggregation efficiency, with a Byzantine adversary detection accuracy of up to 99%.

Predicting the traffic flow has been one of the most important applications in intelligent transportation system. However, the missing information in the traffic data will directly affect the final performance, and it has evolved into a challenge in data analysis based intelligent transportation system applications. Recently, Nuclear Norm-based Tensor Completion algorithm can recover the missing multidimensional information in traffic data recovery by truncated nuclear norm minimization. However, the existing Truncated Nuclear Norm threshold may result in excessive punishment of large singular values, and therefore leading the accurate data missing. To overcome this problem, we present a new Nuclear Norm-based Tensor Completion method, which considers the prior rank information and retains large singular values to approximate the rank of the matrix better. Firstly, to achieve optimized rank parameter, a similar block matrix method is proposed, which takes advantage of the nonlocal similarity to the separate of data and noise. Moreover, an optimal rank estimation algorithm is proposed, which can automatically achieve the truncated threshold parameter through the iterative optimization algorithm. Finally, extensive experiments have been conducted to verify our proposal, and the results show that our method can get better performance in terms of recovery accuracy.

Deep learning has been proved to be one of the most effective method in feature encoding for different intelligent applications such as video-based human action recognition. However, its nonconvex optimization mechanism leads large memory consumption, which hinders its deployment on the smart embedded systems with limited computational resources. To overcome this challenge, we propose a novel deep model compression technique for smart embedded systems, which realizes both the memory size reduction and inference complexity decrease within a small drop of accuracy. First, we propose an improved naive Bayes inference-based channel parameter pruning to obtain a sparse model with higher accuracy. Then, to improve the inference efficiency, the improved Tucker decomposition method is proposed, where an improved genetic algorithm is used to optimize the Tucker ranks. Finally, to elevate the effectiveness of our proposed method, extensive experiments are conducted. The experimental results show that our method can achieve the state-of-the-art performance compared with existing methods in terms of accuracy, parameter compression, and floating-point operations reduction.

Human behavior understanding is playing more and more important role in human-centered Industrial Internet of Video Things (IIoVT) system with the deep combination of artificial intelligence and video-based industrial Internet of Things. However, it requires expensively computational resources, including high-performance computing units and large memory, to train a deep computation model with a large number of parameters, which limits its effectiveness and efficiency for IIoVT applications. In this article, a tensor-train mechanism based deep model is presented for video human behavior understanding to meet the requirement of IIoVT applications. It can get competitive performance in accuracy and training efficiency with potentiality for combination of artificial intelligence and prefront IIoVT system. On the one hand, to achieve desirable accuracy, we improved the conventional CNN and adopted the recurrent neural network mechanism to enhance the video representation over time, which takes the correlation between consecutive deep feature into consideration. On the other hand, to enhance the inference capacity between the spatial and temporal features, we carry out the self-critical reinforcement learning mechanism in parameter learning stage. Meanwhile, to further reduce parameter storage size to meet requirement for the deployment of deep neural network and edge device, the tensor-train mechanism is used, which transforms the parameter matrix to a tensor space and carry out tensor decomposition mechanism to decrease the number of parameter generated in parameter training. Finally, we conduct extensive experiments to evaluate our scheme, and the results demonstrate that our method can improve the training efficiency and save the memory space for the deep computation model with better accuracy.

Video surveillance techniques like scene segmentation are playing an increasingly important role in multimedia Internet-of-Things (IoT) systems. However, existing deep learning-based methods face challenges in both accuracy and memory when deployed on edge computing devices with limited computing resources. To address these challenges, a tensor-train video scene segmentation scheme that compares the local background information in regional scene boundary boxes in adjacent frames is proposed. Compared to the existing methods, the proposed scheme can achieve competitive performance in both segmentation accuracy and parameter compression rate. In detail, first, an improved faster region convolutional neural network (faster-RCNN) model is proposed to recognize and generate a large number of region boxes with foreground and background to achieve boundary boxes. Then, the foreground boxes with sparse objects are removed and the rest are considered as optional background boxes used to measure the similarity between two adjacent frames. Second, to accelerate the training efficiency and reduce memory size, a general and efficient training way using tensor-train decomposition to factor the input-to-hidden weight matrix is proposed. Finally, experiments are conducted to evaluate the performance of the proposed scheme in terms of accuracy and model compression. Our results demonstrate that the proposed model can improve the training efficiency and save the memory space for the deep computation model with good accuracy. This work opens the potential for the use of artificial intelligence methods in edge computing devices for multimedia IoT systems.

Multimedia Internet-of-Things (IoT) systems have been widely utilized in various computer vision tasks and significantly integrated computer vision and networking capabilities. In these systems, convolutional neural networks (CNNs) perform a preliminary analysis of the collected video or image information in the edge devices. However, the high computational cost and huge storage consumption of the complex CNNs prevent their deployment on mobile-edge devices that have limited computational resource and memory. In this article, we aim to simultaneously accelerate and compress CNNs via a multilevel filter pruning (MFP) algorithm, to alleviate the dependence on the hardware of IoT edge nodes. First, a global pruning sensitivity order is defined, which could guide us to perform preliminary pruning from the perspective of convolutional layers' sensitivity. Then, the functional index of each filter is judged by the image entropy of its output feature map, which contributes to further pruning from the perspective of filter function importance. Finally, the moderate fine tuning is adopted to recover the network capability. The experimental results show that the proposed MFP algorithm could reduce 54.5% floating-point operations and 31.9% graphics memory for VGG-16 on CIFAR-10, and achieve 5.45 × floating-point acceleration and 19.70 × storage reduction for VGG-16 on ImageNet. In the reconstruction phase, the algorithm could recover the network capability much faster than the existing pruning algorithms.

Because of the co-existing of H.264/AVC and high efficiency video coding standard (HEVC) in the coming long period, video transcoding technology has become an essential part of multimedia communication in the field of the Internet of Vehicles (IoV). However, due to the huge computational complexity of re-encoding processes, traditionally cascaded transcoders greatly increase the computing burden of the embedded devices and impact the real-time capability of transportation communication systems. In order to address this problem, a fast transcoding solution is proposed in this paper. First, we exploit the mapping relationship among H.264/AVC decoding information and HEVC coding unit (CU) depth decision and prediction unit (PU) mode decision. Then, a three-output classification model is built for CU depth decision processes, and a two-output classification model is built for PU mode selection processes by using support vector machine method. Finally, the models are applied into the cascaded transcoder to accelerate the re-encoding process. The experimental results show that our proposal averagely achieves up to 53.7% and 52.3% complexity reductions under Lowdelay_P_main and Randomaccess_main configurations, respectively, with the negligible rate-distortion degradation, which show a great potential in improving the transcoding efficiency in the real-time video communication system of IoV.