Generative AI (GenAI) has emerged as a transformative technology, enabling customized and personalized AI-generated content (AIGC) services. In this paper, we address challenges of edge-enabled AIGC service provisioning, which remain underexplored in the literature. These services require executing GenAI models with billions of parameters, posing significant obstacles to resource-limited wireless edge. We subsequently introduce the formulation of joint model caching and resource allocation for AIGC services to balance a trade-off between AIGC quality and latency metrics. We obtain mathematical relationships of these metrics with the computational resources required by GenAI models via experimentation. Afterward, we decompose the formulation into a model caching subproblem on a long-timescale and a resource allocation subproblem on a short-timescale. Since the variables to be solved are discrete and continuous, respectively, we leverage a double deep Q-network (DDQN) algorithm to solve the former subproblem and propose a diffusion-based deep deterministic policy gradient (D3PG) algorithm to solve the latter. The proposed D3PG algorithm makes an innovative use of diffusion models as the actor network to determine optimal resource allocation decisions. Consequently, we integrate these two learning methods within the overarching two-timescale deep reinforcement learning (T2DRL) algorithm, the performance of which is studied through comparative numerical simulations.

Low altitude uncrewed aerial vehicles (UAVs) are expected to facilitate the development of aerial-ground integrated intelligent transportation systems and unlocking the potential of the emerging low-altitude economy. However, several critical challenges persist, including the dynamic optimization of network resources and UAV trajectories, limited UAV endurance, and imperfect channel state information (CSI). In this paper, we offer new insights into low-altitude economy networking by exploring intelligent UAV-assisted vehicle-to-everything communication strategies aligned with UAV energy efficiency. Particularly, we formulate an optimization problem of joint channel allocation, power control, and flight altitude adjustment in UAV-assisted vehicular networks. Taking CSI feedback delay into account, our objective is to maximize the vehicle-to-UAV communication sum rate while satisfying the UAV's long-term energy constraint. To this end, we first leverage Lyapunov optimization to decompose the original long-term problem into a series of per-slot deterministic subproblems. We then propose a diffusion-based deep deterministic policy gradient (D3PG) algorithm, which innovatively integrates diffusion models to determine optimal channel allocation, power control, and flight altitude adjustment decisions. Through extensive simulations using real-world vehicle mobility traces, we demonstrate the superior performance of the proposed D3PG algorithm compared to existing benchmark solutions.

With the widespread application of artificial intelligence (AI) in Internet of Vehicles (IoV), particularly in autonomous driving and intelligent traffic management. IoV is facing tremendous pressure for large amounts of data transmission and real-time perception. As a key technology of 6G, integrated sensing and communication (ISAC) is expected to alleviate this pressure by improving spectrum utilization and reducing transmission latency. In addition, when obstacles exist in IoV, communication performance is seriously affected, which is detrimental to driving safety. To address this issue, Reconfigurable Intelligent Surfaces (RIS) as a relay is a feasible solution. Therefore, we construct a RIS-assisted ISAC IoV scenario, and further consider adding dynamic obstacles. Our optimization problem aims to enhance overall performance by maximizing a weighted combination of communication rate and sensing accuracy through the optimization of beamforming and RIS phase shifts. The problem has temporal characteristics and is a Markov Decision Process (MDP). To this end, we use the Long Short-Term Memory-Deep Deterministic Policy Gradient (LSTM-DDPG) algorithm to solve the problem. The results from the simulation illustrate the efficacy of the proposed algorithm in handling dynamic and complex blockage scenarios. Additionally, introducing RIS in blockage scenarios significantly increases the communication rate by up to 58.33%.

Collaborative perception is a vital approach for autonomous systems to understand complex environments by sharing information among multiple agents. However, existing methods often struggle with spatiotemporal misalignment and inefficient feature fusion under dynamic, uncertain conditions. To address these challenges, we propose HGSTA (HyperGraph Spatial-Temporal Awareness), a novel framework that integrates hypergraph computing for multi-scale spatiotemporal feature fusion. By modeling high-order relationships among agents’ data through a hypergraph, HGSTA effectively aligns and combines complementary features from different viewpoints. Extensive experiments on three benchmark datasets demonstrate that HGSTA significantly outperforms current state-of-the-art methods, validating the advantages of our hypergraph-based collaborative perception approach.

With the rapid advancement of artificial intelligence (AI), generative AI (GenAI) has emerged as a transformative tool, enabling customized and personalized AI-generated content (AIGC) services. However, GenAI models with billions of parameters require substantial memory capacity and computational power for deployment and execution, presenting significant challenges to resource-limited edge networks. In this paper, we address the joint model caching and resource allocation problem in GenAI-enabled wireless edge networks. Our objective is to balance the trade-off between delivering high-quality AIGC and minimizing the delay in AI GC service provisioning. To tackle this problem, we employ a deep deterministic policy gradient (DDPG)-based reinforcement learning approach, capable of efficiently determining optimal model caching and resource allocation decisions for AIGC services in response to user mobility and time-varying channel conditions. Numerical results demonstrate that DDPG achieves a higher model hit ratio and provides superior-quality, lower-latency AIGC services compared to other benchmark solutions.

Mobile Crowdsensing (MCS) is a promising paradigm where embedded sensor are exploited for collecting and sharing environmental data. In IoV, participating vehicles sense the environment, collect data and transmit the data to the edge server for processing to provide real-time services. However, real-time services not only pose energy challenges for massive data transmission and analysis, but also demand dynamic multi-timeslot optimisation for MCS systems. Additionally, sensing data often require continuous updates to prevent the provision of obsolete services. To address this, we introduce the concept of Urgency of Information (UoI) to characterize the freshness of sensing data. Diverging from the linear growth trend of Age of Information (AoI), UoI enables dynamic adjustment of the decay rate of data freshness based on traffic complexity. In this paper, we propose a dynamic multi-timeslot MCS system for IoV, under constraints of UoI, which intelligently leverages the spatial correlation of perception to update data and minimizes network energy consumption while ensuring compliance with UoI constraints. Then we propose a Joint Data Collection and Resource Allocation (JDCRA) algorithm to obtain the solution based on convex optimization. To the best of our knowledge, this is the first work to jointly optimize multi-road data collection and resource allocation, taking into account UoI metrics. We evaluate JDCRA by experiments on SUMO in real-world scenarios. Experimental results show that JDCRA outperforms state-of-the-art methods in terms of energy consumption and UoI violation probability, and obtains solutions that consumes only 7.43% more energy than the optimal solution in polynomial complexity.

Semantic Communication (SemCom) has attracted significant attentions due to its potential to enhance communication efficiency and support human-centric services in 6G networks. However, the presence of mismatched background knowledge and dynamic communication channels decreases the performance of SemCom. These issues ultimately lead to a degradation in users’ quality of experience (QoE). To overcome this challenge, a hybrid semantic and bit communication framework is proposed to effectively improve communication performance under mismatched knowledge constraints. Specifically, we design a time division duplex (TDD) SemCom scheme, where the transmitter and the receiver synchronize background knowledge through the uplink transmission to eliminate mismatch constraints. To guide subframe configuration and communication mode selection in the TDD system, a novel QoE model including perceived quality and energy consumption is proposed, and a long-term average QoE maximization problem is further formulated. To solve the proposed NP-hard problem, a joint subframe configuration and communication mode selection algorithm (JSCA) is designed, and the original problem is decomposed into two subproblems. Firstly, the subframe configuration subproblem is transformed into a quasi-concave problem, and the optimal solution is obtained by the bisection method. Secondly, a deep reinforcement learning (DRL)-based approach is designed to select the communication mode for each service. The numerical results validate the effectiveness of JSCA and demonstrate that the proposed hybrid semantic and bit communication scheme can achieve higher QoE compared with fixed schemes, especially in long-term service scenarios.

The rapid advancement of Artificial Intelligence (AI) has introduced Deep Neural Network (DNN)-based tasks to the ecosystem of vehicular networks. These tasks are often computation-intensive, requiring substantial computation resources, which are beyond the capability of a single vehicle. To address this challenge, Vehicular Edge Computing (VEC) has emerged as a solution, offering computing services for DNN-based tasks through resource pooling via Vehicle-to-Vehicle/Infrastructure (V2V/V2I) communications. In this paper, we formulate the problem of joint DNN partitioning, task offloading, and resource allocation in VEC as a dynamic long-term optimization. Our objective is to minimize the DNN-based task completion time while guaranteeing the system stability over time. To this end, we first leverage a Lyapunov optimization technique to decouple the original long-term optimization with stability constraints into a per-slot deterministic problem. Afterwards, we propose a Multi-Agent Diffusion-based Deep Reinforcement Learning (MAD2RL) algorithm, incorporating the innovative use of diffusion models to determine the optimal DNN partitioning and task offloading decisions. Furthermore, we integrate convex optimization techniques into MAD2RL as a subroutine to allocate computation resources, enhancing the learning efficiency. Through simulations under real-world movement traces of vehicles, we demonstrate the superior performance of our proposed algorithm compared to existing benchmark solutions.

The Vehicular Fog Computing (VFC) provides low-latency computing service to support emerging intelligent transportation applications in Internet of Vehicles (IoV). Multi-node cooperative VFC can utilize Connected and Autonomous Vehicles (CAVs) to implement cooperative intelligence. Due to the mobility of vehicles, service migration is necessary when task offloading service providers change. This paper proposes an Asynchronous Task Offloading Scheme (ATO-S) that allows each CAV to choose an independent optimization period and provides robust task offloading services under unknown vehicle mobility probability distribution. To the best of our knowledge, this is the first work to investigate asynchronous and robust multi-node cooperative task offloading in dynamic VFC-enhanced IoV scenarios. Furthermore, we formulate the long-term energy consumption minimization problem of VFC and transfer it into each time slot problem by Lyapunov optimization. Then we design Asynchronous Task Offloading Algorithm (ATO-A) to jointly optimizing CAVs matching, communication and computation resource allocation, and transmission power based on multiple mathematical techniques and hybrid heuristic algorithm. Extensive simulations based on real-world traffic scenario are conducted by varying multiple crucial parameters. Simulation results demonstrate the energy efficiency and task queue stability achieved by ATO-A, and service robustness achieved by ATO-S, in comparison with benchmark solutions.

Task scheduling as an effective strategy can improve application performance on computing resource-limited devices over distributed networks. However, existing evaluation mechanisms for application completion fail to depict the complexity of diverse applications and time-varying networks, which involve dependencies among tasks, computing resource requirements, multi-dimensional quality of service (QoS) constraints, and limited contact duration among devices. Furthermore, traditional QoS-oriented task scheduling strategies struggle to meet the performance requirements without considering differences in satisfaction and acceptance of the application, leading to application failures and resource wastage. To tackle these issues, a quality of experience (QoE) cost model is designed to evaluate application completion, depicting the relationship among application satisfaction, communications, and computing resources over the time-varying distributed networks. Specifically, considering the sensitivity and preference of QoS, we model the different dimensional QoS degradation cost functions for dependent tasks, which are then integrated into the QoE cost model. Based on the QoE model, the dependent task scheduling problem is formulated as the minimization of overall QoE cost, aiming to improve the application performance over the time-varying distributed networks, which is proven Np-hard. Moreover, a heuristic Hierarchical Multi-queue Task Scheduling (HMTS) algorithm is proposed to address the QoE-oriented task scheduling problem among multiple dependent tasks, which utilizes hierarchical multiple queues to determine the optimal task execution order and location according to different dimensional QoS priorities. Finally, extensive experiments demonstrate that the proposed algorithm can significantly improve the satisfaction of applications.

Face anti-spoofing (FAS) is crucial for enhancing the security of face recognition systems. Although previous multi-frame face anti-spoofing methods utilized consecutive images for decision fusion, they lacked the extraction and detection of facial behavior, which is an essential attribute to distinguish real faces from presentation attacks (PAs). Besides, people tend to display unconscious facial behaviors at specific moments, other intervals are often redundant for classification. To address these issues, in this paper, we propose the frame-level-attention-based facial behavior mining for FAS (FBM-FA). An adaptive key-frame mining module is introduced to locate potential peak frames of facial behavior in a video. Moreover, a two-stream feature extraction network is designed to extract temporal features between facial behaviors and single facial behavior representations from position information. Comprehensive experiments are performed on four datasets including OULU-NPU, 3DMAD, CASIA-FASD, and REPLAY-ATTACK. Experiments demonstrate that our method can achieve perfect detection performance on 2D and 3D datasets.

It’s essential to effectively learn the spatial-temporal information of point cloud sequences for 3D action recognition, especially for the increasing popular application of Human Action Recognition (HAR) at smart home. In this paper, we utilize Frequency Modulated Continuous Wave (FMCW) radar to acquire point clouds and design the end-to-end network which can learn the spatial-temporal representation of dynamic 3D point cloud sequences, dubbed Spatial and Temporal Point Network (STPointNet). STPointNet is a permutation-invariant network that can learn from unstructured point clouds with irregular domains. The proposed STPointNet network utilizes the Spatial Part to extract 1024-dimensional global spatial features of point clouds and the Temporal Part to extract 1024-dimensional global temporal features of point clouds. These features are then concatenated to form a complete feature vector of 2048 dimensions, which is subsequently fed into a Multi-Layer Perceptron (MLP) with a non-linear activation function softmax to obtain classification scores. Extensive experiments show that the proposed method significantly outperforms existing state-of-the-art approaches according to the recognition accuracy, achieving a recognition accuracy of 98.26% on the MMAction dataset. Furthermore, it demonstrates excellence and robustness in utilizing sparse point clouds for 3D action recognition.

Anomaly detection based on key performance indicator (KPI) is an important topic in the field of intelligent operation and maintenance. The problem of insufficient annotated samples is widespread in the industrial Internet, and it severely impairs the performance of data-driven anomaly detection. Previous methods tackled the problem mainly through unsupervised methods, which rely heavily on the ability of the algorithm to extract features. In this work, we propose an unsupervised hybrid model to address these issues. Technically, we capture the long-term dependencies of time series by stacked BiLSTM and use the self-attention mechanism to adaptively select the most noteworthy information in the time series data for global consideration. To effectively distinguish anomalies, we propose a dynamic threshold method that takes into account the context of the data being measured. By doing so, we aim to minimize false positives and missed positives, thus significantly enhancing the overall performance. Extensive experiments on various public benchmarks and real-world measured data demonstrate that our method offers advanced performance and practicality.

Autonomous driving research has predominantly focused on LiDAR and vision-based methodologies. While LiDAR excels in accuracy and robustness, its high cost is prohibitive; vision-based systems, alternatively, are more economical but limited in scope and precision. To overcome these limitations, this paper presents a cross-modal scene recognition algorithm integrating semantic information to facilitate a seamless positional transformation between vision devices and LiDAR maps. The core objective is to enable precise initial localization within LiDAR point cloud maps, thereby establishing a consistent linkage between visual perception and spatial mapping. The algorithm utilizes a cross-modal interaction network to synergize features from both modalities, significantly narrowing the semantic gap. Further enhancing this framework, graph neural networks are employed to deepen the semantic understanding and improve alignment between disparate modal scenes. This method demonstrates remarkable efficiency in decoding complex environmental contexts and elevating match precision. Validated on the KITTI dataset, the algorithm achieved a commendable average F1 score of 0.815, affirming its value in advancing autonomous navigation systems with more accurate and reliable scene recognition capabilities.

With the rapid development of vehicular networks, ever-growing number of on-board sensors makes vehicle applications/tasks to be not only computation-intensive but also data-intensive. To this end, Vehicular Cloud Computing (VCC) has been convinced as a promising paradigm to offer valuable computing and sensing services to vehicles. However, considering the heterogeneity of on-board computation and sensing capabilities, how to efficiently determine the appropriate vehicle to process the task is challenging. Also, the allocation of transmission power can significantly impact the corresponding energy consumption. Therefore, to minimize the weighted sum of execution delay and energy consumption of vehicle tasks, in this paper, we propose a Collaborative Sensing-Assisted Task Offloading and Resource Al-location (CSTR) algorithm based on the Integrated Sensing and Communication (ISAC) mechanism. The optimization problem is formulated as a mixed integer nonlinear programming problem (MINLP), which is proven to be NP-hard. To achieve this, the original problem is decoupled into two sub-problems namely the task offloading problem and transmission power allocation problem, which are solved by Genetic Algorithm (GA) and convex optimization technique, respectively. Validation through several simulations based on real-world road networks has demonstrated that our proposed CSTR can outperform existing benchmark solutions under various settings.

In the development of Intelligent Transportation Systems (ITS) and the Internet of Vehicles (IoV), traditional management strategies that separate communication, sensing, and computation resources, as well as the uneven scheduling of resources, are increasingly unable to meet the needs of complex scenarios such as higher-level autonomous driving and integrated vehicle-road-cloud systems. This paper proposes a Quality of Service (QoS) tripartite evolutionary game model based on task-driven Integrated sensing, communication, and computation (ISCC). The dynamic evolution of resource allocation is analyzed by constructing a resource game platform centered on task requirements. Additionally, a joint utility function involving multi-node competition, dependency, and cooperation as QoS indicators has also been developed to achieve a unified representation of performance metrics. Subsequently, task-driven analysis effectively guides resource allocation strategies, ensuring optimization of resource utilization and precise alignment with task requirements. This paper offers new methodological support for designing and optimizing future intelligent transportation systems.

Active reconfigurable intelligent surface (RIS) acts as an enhancement of signal transmitted by the base station (BS) to overcome the effects of multiplicative fading and provide communication services for users. In this paper, we investigate an active RIS-assisted integrated sensing and communication (ISAC) system in which the BS senses the target in the presence of interference and establishes communication with multiple users assisted by the RIS. The optimization problem is formulated to jointly design the BS’s sensing transceiver beamforming, communication beamforming, and the RIS’s reflection coefficients to maximize the sensing SINR while satisfying the communication SINR requirements of users. An algorithm based on the ideas of quadratic transformation (QT), Karush-Kuhn-Tucker (KKT) conditions, semidefinite relaxation (SDR), and alternating optimization (AO) is designed to transform and solve the non-convex problem optimally. Simulation results verify the effectiveness and superiority of active RIS-assisted ISAC systems in improving sensing performance with communication SINR constraints.

Abstract Beam tracking is commonly employed in millimetre wave (mmWave) based vehicle‐to‐vehicle (V2V) networks to align the beams towards the intended targets and compensate for the path loss of mmWave signals. To mitigate the high latency issue arising from the tracking processes, integrated sensing and communication (ISAC) technology leverages the echo signal to sense the motion parameters of the target, achieving low‐latency beam tracking without requiring pilot and uplink feedback. Existing studies mainly focus on utilizing ISAC for beam alignment to track the target, without integrating beam tracking with resource allocation. In this paper, we propose the ISAC beam tracking based on deep reinforcement learning (IBTD) algorithm to address this problem. Specifically, we introduce the concept of packet age to measure communication performance. To achieve accurate beam tracking and optimize the transmit power, we integrate the sensing results, such as the position and velocity of the target vehicle, along with the buffer pool status information, with deep reinforcement learning (DRL) to select an appropriate policy. Furthermore, we consider the effect of inter‐vehicle distance and incorporate the changing of tracking targets into the DRL‐based policy. Simulation results demonstrate that the proposed IBTD algorithm achieves lower packet age and transmit power consumption compared to the baseline algorithms.

Unmanned aerial vehicles (UAVs) have attracted massive attentions in Internet of Vehicles (IoV) networks to support the communications among roadside units (RSUs) and IoV users. In UAV-assisted IoV systems, UAVs and RSUs generally work within the same frequency band to improve the system spectral utilization efficiency, limited by the lack of spectrum resources, which, however, can cause mutual interference. To cope with the interference and increase the capacity of UAV-assisted IoV systems, this article considers to distribute part of the data traffic from the ground IoV system to the unlicensed spectrum. Specifically, we consider a heterogeneous communication scenario, in which a UAV-assisted IoV system and a Wi-Fi system coexist well: the RSUs can properly occupy unlicensed spectrum to increase the capacity of the UAV-assisted IoV system while mitigating interference between the UAVs and RSUs, without affecting the Wi-Fi system’s communication performance. We then propose a joint power control and time allocation scheme for the UAV-assisted IoV system over licensed and unlicensed spectrum. Joint optimization method is used to obtain the optimal power and time allocation strategy to maximize the overall system capacity. Simulation results and comprehensive analysis have demonstrated the superior performance of the proposed scheme, as compared to the conventional and state-of-art resource allocation strategies.

Vehicular Clouds (VCs) are modern platforms for processing of computation-intensive tasks over vehicles. Such tasks are often represented as Directed Acyclic Graphs (DAGs) consisting of interdependent vertices/subtasks and directed edges. However, efficient scheduling of DAG tasks over VCs presents significant challenges, mainly due to the dynamic service provisioning of vehicles within VCs and non-Euclidean representation of DAG tasks’ topologies. In this paper, we propose a Graph neural network-Augmented Deep Reinforcement Learning scheme (GA-DRL) for the timely scheduling of DAG tasks over dynamic VCs. In doing so, we first model the VC-assisted DAG task scheduling as a Markov decision process. We then adopt a multi-head Graph ATtention network (GAT) to extract the features of DAG subtasks. Our developed GAT enables a two-way aggregation of the topological information in a DAG task by simultaneously considering predecessors and successors of each subtask. We further introduce non-uniform DAG neighborhood sampling through codifying the scheduling priority of different subtasks, which makes our developed GAT generalizable to completely unseen DAG task topologies. Finally, we augment GAT into a double deep Q-network learning module to conduct subtask-to-vehicle assignment according to the extracted features of subtasks, while considering the dynamics and heterogeneity of the vehicles in VCs. Through simulating various DAG tasks under real-world movement traces of vehicles, we demonstrate that GA-DRL outperforms existing benchmarks in terms of DAG task completion time.

Federated learning (FL) is a prominent paradigm of 6G edge intelligence (EI), which mitigates privacy breaches and high communication pressure caused by conventional centralized model training in the artificial intelligence of things (AIoT). The execution of multimodal federated perception (MFP) services comprises three sub-processes, including sensing-based multimodal data generation, communication-based model transmission, and computing-based model training, ultimately competitive on available underlying multi-domain physical resources such as time, frequency, and computing power. How to reasonably coordinate the multi-domain resources scheduling among sensing, communication, and computing, therefore, is vital to the MFP networks. To address the above issues, this article explores service-oriented resource management with integrated sensing, communication, and computing (ISCC). Specifically, employing the incentive mechanism of the MFP service market, the resources management problem is defined as a social welfare maximization problem, where the concept of “expanding resources” and “reducing costs” is used to enhance learning performance gain and reduce resource costs. Experimental results demonstrate the effectiveness and robustness of the proposed resource scheduling mechanisms.

Abstract The future vehicle‐to‐everything (V2X) scenarios greatly call for reliable and efficient communication links between roadside units (RSUs) and vehicles. Integrated sensing and communication (ISAC) represents one of the key candidate technologies for 6G, which provides new solutions for this problem. In this paper, the beam tracking design for V2X communication is investigated by exploiting the ISAC technique, where the RSU uses the echoes of the ISAC signals to estimate the motion parameters of the vehicles. Compared with traditional feedback‐based beam tracking, the proposed method can reduce the overall signaling costs and improve communication robustness. Furthermore, to achieve high accuracy and low latency for beam alignment, a Trajectory Adaptive Unscented Kalman Filter (TA‐UKF) beam tracking algorithm is proposed, which obtains the optimal state transition model following the change of the vehicle state. Considering practical application scenarios, an ISAC‐based adaptive beam coverage (ISAC‐AC) scheme is further developed for the beam tracking of extended targets, which can achieve the beam misalignment of vehicles in practical scenarios. Numerical results demonstrate that the proposed beam tracking and coverage strategies outperform baseline approaches.

Implementing either Federated learning (FL) or split learning (SL) over clients with limited computation/communication resources faces challenges on achieving delay-efficient model training. To overcome such challenges, we investigate a novel distributedClustering-basedHybrid fEdEratedSplit lEarning (CHEESE) framework, consolidating distributed resources among clients by device-to-device (D2D) communications, working in an intra-serial inter-parallel manner. InCHEESE, each learning client can form a cluster with its neighboring helping clients via D2D communications to train an FL model collaboratively. Inside each cluster, the model is split into multiple segments via a model splitting and allocation (MSA) strategy, while each cluster member trains one segment. After completing intra-cluster training, a transmission client (TC) is determined from each cluster to upload a complete model to the base station for global model aggregation under allocated bandwidth. Accordingly, an overall training delay cost minimization problem is formulated, involving the following subproblems: client clustering, MSA, TC selection, and bandwidth allocation. Due to its NP-Hardness, the problem is decoupled and solved iteratively. The client clustering problem is first transformed into a distributed clustering game based on potential game theory, where each cluster further investigates the remaining three subproblems to evaluate the utility of each clustering strategy. Specifically, a heuristic algorithm is proposed to solve the MSA problem under a given clustering strategy, while a greedy-based convex optimization approach is introduced to solve the joint TC selection and bandwidth allocation problem. Extensive experiments on practical models and datasets demonstrate thatCHEESEcan significantly reduce training delay costs.

Most existing face anti-spoofing methods get excellent results in the intra-dataset experiments. However, in the cross-dataset experiments, the performance may decline sharply due to the change of lighting condition, spoofing type, camera and other factors. To solve this issue, many researchers have proposed domain generalization methods for face anti-spoofing task. However, most approaches simply take full face image as input, thus the model may focus on a fixed area of the face. So these methods still have poor generalization ability to local spoofing attacks (as shown in Fig. 1). In this paper, we propose a dynamic attention based domain generalization method for face anti-spoofing. Specifically, the proposed method takes both of the full face and randomly selected local region as inputs. Moreover, we discover that the performance of the network changes with the position of local input. To handle this problem, we utilize the designed Gaussian function to fit the changing pattern of performance, and apply it into weighted cross-entropy loss function. Moreover, we introduce a Feature Decoupling Module (FDM) that can decouple features into domain-invariant and domain-specific features. In order to decouple correctly, we propose a Style Loss function to constrain the domain-invariant features of different samples aggregated, while domain-specific features separated. Extensive experiments are conducted to show the superior generalization ability of the proposed method.

Modern communication and computing technology is the basic support of the industrial intelligent systems (IIS). As a key component of IIS, the smart port is essential to be offered low-complexity and high-reliability communication service, especially for driverless engineering vehicles. However, it is combined and nonconvex to find the optimal association between vehicles and the road side units (RSUs). Besides, due to the mobility of vehicles and the severe path loss of mmWave links, beam switching and reassociation between vehicles and RSUs are required frequently, which brings a great challenge to the communication for the IIS. A low complexity closed-loop strategy based on distributed cooperation for mmWave communication in IIS is proposed in this study, in which user association and beam tracking with the assistance of beam pools is proposed. Many-to-many user association is established based on distributed multiagent reinforcement learning, where the vehicle can independently select the set of serving RSUs based on the local observation without information exchange with others, reducing the signaling overhead and computational complexity while improving system throughput. Furthermore, multipoint-cooperation soft switching of beams based on beam tracking improves the reliability of mmWave communication with the smaller training cost. Extensive analysis and simulation results demonstrate that the proposed solution significantly reduces the complexity of the mmWave communication while improving the throughput and stability in IIS.

Fog computing is an energy-efficient and cost-effective paradigm to help alleviate the pressure of resource-constrained mobile devices (MDs) running computation-intensive applications. In this article, we investigate the joint task partitioning and power control problem in a fog computing network with multiple MDs and fog devices (FDs), where each MD has to complete a periodic computation task under the constraints of delay and energy consumption. Each task can be partitioned into multiple subtasks and offloaded to the FDs according to the task partition strategy and transmission power strategy to reduce task execution delay and energy consumption. To this end, we present a multiagent deep deterministic policy gradient (MADDPG)-based task offloading algorithm for MDs to maximize the long-term system utility including the execution delay and energy consumption. Each MD inputs the local information, e.g., the task requirements, the available communication, and computation resources of the FDs, the computation resources, and the battery level of the MD into a distributed actor network to generate a task offloading policy, while a centralized critic network is used to update the weights of the actor networks to improve offloading performance. Numerical simulation results demonstrate the effectiveness of the proposed scheme in improving the system utility, reducing the average execution delay as well as the average energy consumption.

In this paper, we investigate the multi-task federated learning in mobile edge computing (MEC) networks where a central server assigns different federated learning tasks to different MEC servers and select feasible clients to participate in the federated learning training process. The problem is formulated as a joint client selection and task assignment problem to maximize the total utility of all tasks, subject to the trained model quality and total training latency. Since the above-mentioned problem is NP-Hard, it poses challenges to obtain the optimal solution within polynomial time, the problem is transformed into a many-to-one-to-one 3D matching problem. To further reduce the computation while ensuring the matching stability, we first adopt the spectral clustering algorithm to cluster the clients into multiple client clusters. Then we reformulate the problem as a 3-Partite weighted hypergraph total weight maximization problem. Finally, we propose a greedy and local search (GLS) based algorithm to resolve the problem. Simulation results demonstrate the effectiveness of the proposed algorithm as compared with baseline schemes.

This paper considers unmanned aerial vehicle (UAV)-assisted infrastructure-to-vehicle (I2V) communication employing UAVs as relays to increase the throughput between a roadside unit (RSU) and a vehicular user equipment (VUE). We investigate the UAV position planning problem under both single UAV and multiple cooperative UAVs scenarios while considering the mobility of the VUE, aiming to maximize the data rate of the system. We first consider using a single UAV and prove that the single UAV position planning can be formulated as a convex optimization problem, and then obtain the optimal position of the UAV. Next, we investigate the multiple cooperative UAVs scenario and formulate the joint power control and position planning problem to improve the data rate of the system under a fixed total power consumption. Numerical simulations are provided to verify our theoretical results. Our findings highlight the effects of important system parameters, such as height, transmit power, and the number of UAVs, on the optimal UAV positioning and system performance.

Visual Simultaneous Localization and Mapping (SLAM) is widely employed in modern mobile service robots, which help robots to capture images indoors for estimating their pose. However, as part of visual SLAM, the typical visual odometry (VO) and visual-inertial odometry (VIO) systems only work in static environments. In many scenarios, they have to work in high-dynamic environments, which brings challenges to previous visual SLAM. In this paper, we proposed a novel monocular VIO for the challenging dynamic environments. Our method can make the robot locate accurately and robustly. Based on VINS-Mono, our system constructs a dynamic objects and feature points detection module. This module combines semantic object detection, multi-sensor-aided, and the geometric 3D vision constraints to remove the dynamic feature points. According to our experiments, the results demonstrate our system outperforms SOTA monocular VIO systems in accuracy and robustness, especially in high-dynamic indoor environments.

Driven by the tremendous in vehicular networks computation-intensive application demands, the incorporation of mobile edge computing (MEC) and vehicular cloud is convinced as a promising paradigm to fulfill computation offloading requirements. However, the changing vehicular communication topology (CVCT) poses a significant challenge for offloading directed acyclic graph (DAG) model application. Due to the precedence and connection constraint between different sub-jobs, the successful offloading of DAG-enabled apllication will be disturbed even interrupted without considering CVCT. To address this problem, we propose a topology-aware dynamic computaion offloading mechanism and adopt simulated annealing algorithm (TASA) to jointly optimize the energy consumption and completion time under dynamic environment, while guaranteeing the convergence of the proposed method. Simulation results reveal the effectiveness of the proposed method in overcoming CVCT’s influence.

Internet of Underwater Things (IoUT) consists of numerous sensor nodes distributed in an underwater area for sensing, collecting, processing information, and sending related messages to the data processing center. However, the characteristics of the underwater environment will bring strict limitations on communication coverage and power scarcity to IoUT networks. Applying cooperative communications to IoUT networks can expand the communication range and alleviate power shortages. In this article, we investigate the cooperative communication problem in a power-limited cooperative IoUT system and propose a reinforcement learning-based underwater relay selection strategy. Specifically, we first determine the optimal transmit powers of the source node and the selected underwater relay to maximize the end-to-end signal-to-noise ratio of the system. Then, we formulate the underwater cooperative relaying process as a Markov process and apply reinforcement learning to obtain an effective underwater relay selection strategy. The simulation results show that the performance of the proposed scheme outperforms that of the equal transmit power settings under the same conditions. In addition, the proposed deep Q-network-based underwater relay selection strategy improves the communication efficiency compared with the Q-learning-based strategy, and the number of iterations needed for convergence can be effectively reduced.

Mobile edge computing (MEC) enabled user-centric ultra-dense network (UDN) is a promising solution to the energy constrained mobile users with delay-sensitive and computation intensive applications. Due to the high density of access points and MEC servers in UDN, both task offloading decision, power control, communication and computation resource allocation need to be addressed. To this end, we consider the joint problem of task offloading, uplink transmission power control, communication and computation resource allocation in a UDN, where the task of each user can be partitioned into several subtasks and offloaded to different access points. To handle the continuous action space of task partitioning and power control, we propose a multi-agent deep deterministic policy gradient (MADDPG) approach to solve this problem. Simulation results reveal the effectiveness of the proposed method.

With the increasing demand of vehicular data transmission, the utilization of cellular resources in low frequency bands is facing great challenges to meet the growing throughput requirements of cellular vehicle-to-everything (C-V2X) users. To solve this problem, we expand certain aspects of the vehicular business to the unlicensed spectrum, which enables C-V2X users to access unlicensed channels fairly and thus will greatly increase system capacity. Moreover, this approach also introduces coexistence issues between C-V2X users and unlicensed users. In this paper, a C-V2X and Wi-Fi coexistence scheme based on reinforcement learning is proposed while considering the system throughput and fairness. A Q-learning algorithm is utilized to determine the optimal duty cycle selection strategy in a multi-unlicensed-channels scenario. Simulation results show that compared with existing coexistence schemes, the proposed scheme can improve throughput performance considerably while ensuring fairness.

With the advantages of Millimeter wave in wireless communication network, the coverage radius and inter-site distance can be further reduced, the ultra dense network (UDN) becomes the mainstream of future networks. The main challenge faced by UDN is the serious inter-site interference, which needs to be carefully addressed by joint user association and resource allocation methods. In this paper, we propose a multi-agent Q-learning based method to jointly optimize the user association and resource allocation in UDN. The deep Q-network is applied to guarantee the convergence of the proposed method. Simulation results reveal the effectiveness of the proposed method and different performances under different simulation parameters are evaluated.

The heterogeneous vehicular network (HetVNET), which consists of multiple radio access networks (RANs), is a promising paradigm to provide a variety of services on the road. However, with the diverse quality of experience (QoE) requirements for vehicles, how to choose the optimal network for the vehicles pose great challenges. In this paper, we investigate the network selection problem in a HetVNET, which includes LTE-vehicle-to-anything (LTE-V2X), dedicated short-range communications (DSRC), WiFi. The network selection problem is formulated as a stable matching between the vehicles and different RANs. A two-sided one-to-many matching algorithm is presented based on the preference lists of both vehicles and RANs. Numerical simulation results show that the proposed method can improve the total throughput of the system and reduce the total network switch time compared to some existing algorithms.

Graph jobs represent a wide variety of computation-intensive tasks in which computations are represented by graphs consisting of components (denoting either data sources or data processing) and edges (corresponding to data flows between the components). Recent years have witnessed dramatic growth in smart vehicles and computation-intensive graph jobs, which pose new challenges to the provision of efficient services related to the Internet of Vehicles. Fortunately, vehicular clouds (VCs) formed by a collection of vehicles, which allows jobs to be offloaded among vehicles, can substantially alleviate heavy onboard workloads and enable on-demand provisioning of computational resources. In this article, we present a novel framework for VCs that maps components of graph jobs to service providers via opportunistic vehicle-to-vehicle communication. Then, graph job allocation over VCs is formulated as a nonlinear integer programming with respect to vehicles' contact duration and available resources, aiming to minimize the job completion time and data exchange cost. The problem is addressed for two scenarios: 1) low-traffic and 2) rush-hour scenarios. For the former, we determine the optimal solutions for the problem. In the latter case, given the intractable computations for deriving feasible allocations, we propose a novel low complexity randomized graph job allocation mechanism by considering hierarchical tree-based subgraph isomorphism extraction. The evaluation of the performance of both optimal and proposed randomized algorithms with two greedy-based baseline methods is carried out through extensive simulations.

In recent years, virtual/augmented reality (VR/AR) technology has received great attention due to its capability of creating various levels of immersive experiences. However, current wireless VR/AR devices are quite expensive, which hinders its large-scale deployment in practice. In this demo, we present a wireless interactive VR/AR teaching system based on popular Android phones. In such a demo, when a teacher explains a 3D model, multiple students can see it from exactly the same perspective as the teacher does through VR/AR glasses. When one student has a concern or question regarding a particular part of the 3D model, he/she can point it out, and a corresponding blue cursor will appear on screens of all users. Moreover, in the absence of 3D models in Android phones, we broadcast 3D models based on their visual priorities.

Device-to-device (D2D) communications provide a promising technique for the fifth-generation (5G) wireless networks. It has been considered to support intelligent vehicular communications, such as in LTE-Vehicle (LTE-V). Particularly, it is very challenging to manage simultaneous handover of a massive number of fast-moving devices on the cell edge. In this paper, we propose a joint handover approach that effectively exploits D2D multicast and D2D clusters to address the severe resource contention caused by the high density and high mobility of mobile devices. The handover decision takes into account the received signal strength (RSS) from each candidate base station (BS) and the moving speed toward each BS. The simulation results show that the proposed solution can significantly reduce the communication interruption probability and handover latency of mobile devices in high-density and high-speed scenarios.

Adversary unmanned aerial vehicles (UAVs) seriously threaten public security and user privacy. In this paper, we propose a reinforcement learning (RL) based defense framework to address malicious UAVs close to a target estate such as a company or an institute. This framework uses Q-learning to choose the defense policy such as jamming the global positioning system signals (GPS) and hacking, and laser shooting. According to the defense history and the current security status of the target estate, this scheme can improve the UAV defense performance in the dynamic game without being aware of the UAV attack policy and environment model in the area of interests. Simulation results show that this scheme can reduce the risk rate of the estate and improve the utility compared with the benchmark scheme against malicious UAVs.

With the rapid development of science and technology, unmanned aerial vehicles (UAVs) gradually become the worldwide focus of science and technology. Not only the development and application but also the security of UAV is of great significance to modern society. Different from methods using radar, optical or acoustic sensors to detect UAV, this paper proposes a novel distance-based support vector data description (SVDD) algorithm using hash fingerprint as feature. This algorithm does not need large number of training samples and its computation complexity is low. Hash fingerprint is generated by extracting features of signal preamble waveforms. Distance-based SVDD algorithm is employed to efficiently detect and recognize low, slow, small unmanned aerial vehicles (LSSUAVs) using 2.4GHz frequency band.

In cognitive radio networks, cooperative spectrum sensing(SS) between cognitive users can improve the detection performance, reduce the testing time. However, a large amount of data interaction restrict its application and increase the consumption of resources. For the demand of cooperative SS, this paper improved the compression algorithm based energy detection which in the early stage of the work, the spectrum data is divided into segments of different characteristic and respectively compressed. The presented algorithm further improved the compression ratio compared to the original algorithm. Verified by the experiment of satellite signals, the experimental results show that the compression performance of this algorithm will be increased several times compared to JPEG, JPEG2000 and detection-based compression algorithm.

This paper proposes a full-duplex physical security model with self-interference remaining. This model doesn't need the assistance of external jamming nodes and it ensures that the uplink and downlink transmission of the full-duplex system can achieve the required secrecy rate. Meanwhile, it creates a base station transmission power optimization problem with flexible constraints and a two-level method to achieve the optimization, so that power optimization can take place in the model with self-interference remaining.

Device-to-device (D2D) communication, which is defined as a direct communication between two mobile users without traversing the Base Station (BS) or the core network to offload the increasing traffic to the user equipments, is one of the key technologies in the fifth generation (5G) of wireless communication systems. Discovery and communication are the basic two features to fulfill the need for the D2D communication. However, Most of existing D2D studies focused on the communication issues always assume that the discovery is completed. In this paper, we propose two strategies of device discovery and access scheme for the 5G cellular networks. Then the performance analysis based on two dimensional discrete time Markov process model is provided. In addition, we present numerical simulation on the Matlab platform. The simulation results demonstrate the viability of the proposed scheme.

Abstract A large number of new data‐consuming applications are emerging, and many of them involve mobile users. In the next generation of wireless communication systems, device‐to‐device (D2D) communication is introduced as a new paradigm to offload the increasing traffic to the user equipment. Before the traffic transmission, D2D discovery and access procedure is the first important step which needs to be completed. In this paper, our goal is to design a device discovery and access scheme for the fifth generation cellular networks. We first present two types of device discovery and access procedures. Then we provide performance analysis based on the Markov process model. In addition, we present numerical simulation on the Vienna Matlab platform. The simulation results demonstrate the viability of the proposed scheme. Copyright © 2015 John Wiley & Sons, Ltd.

Cognitive radio networks (CRNs) are threatened by smart jammers that aim to block the ongoing transmissions of secondary users according to their transmission patterns obtained from public control channels or compromised secondary users. Without requiring any pre-shared PHY-layer keys at the receivers, the uncoordinated frequency hopping (UFH) technique that was proposed to address smart jammers suffers from a low communication efficiency. In this paper, we develop a UFHbased collaborative broadcast system over universal software radio peripherals (USRPs), which exploits the node cooperation to improve the broadcast efficiency and jamming resistance of CRNs. Experiments are performed over USRPs to evaluate the broadcast performance against smart jammers under various network topologies. We investigate the impact of the CRN bandwidth, the prediction accuracy of smart jammers regarding the CRN frequency hopping pattern, and the jamming power and signal pattern. Experimental results show that the proposed broadcast system is more robust than three benchmark broadcast systems in most jamming scenarios.

ABSTRACT The IEEE 802.16 mesh network is a promising next generation wireless backbone network. In the network, the allocation of minislots is handled by centralized scheduling and distributed scheduling, which are independently exercised. However, the standard does not specify how the frame can be partitioned among its centralized and distributed schedulers. Through efficient partitioning that dynamically adapts the partitioning based on demand, network can support more user applications. Although a dynamic frame partitioning scheme to use Markov model has been studied, the dynamic frame partitioning method has not been fully investigated. This paper proposes two novel and general dynamic frame partitioning scheme for IEEE 802.16 mesh networks so that the minislot allocation can be more flexible and the utilization is increased. The two schemes respectively use GM(1,1)‐Markov model and Grey–Verhulst–Markov model to predict efficient partitions for future frames according to the minislot utilization in current frames. Our study indicates that the two proposed schemes outperform the scheme of using Markov model. Copyright © 2012 John Wiley & Sons, Ltd.

ABSTRACT This paper proposes a time–frequency‐based fingerprinting identification approach by extracting the transient characterizations regarding highly integrated wireless local area network (WLAN) cards. The transient energy envelope is derived from the slice of spectrogram in the time–frequency domain. Then this transient response is fitted to a polynomial under least square criteria, and the polynomial coefficients are regarded as the feature vector. A data acquisition system has been set up to capture IEEE 802.11b Wi‐Fi (wireless fidelity) signals. The results exhibit an approving distinctiveness of up to 94% to classify different manufactories of WLAN cards and of 78.4% for the WLAN cards of the same manufactory. Copyright © 2011 John Wiley & Sons, Ltd.

In sensor networks, due to power outage at a sensor node, hardware dysfunction, or bad environmental conditions, not all sensor samples can be successfully gathered at the sink. Additionally, in the data stream scenario, some nodes may continually miss samples for a period of time. In this paper, a sparsity-based online data recovery approach is proposed. We construct an over complete dictionary composed of past data frames and traditional fixed transform bases. Assuming the current frame can be sparsely represented using only a few elements of the dictionary, missing samples in each frame can be estimated by Basis Pursuit. Our method was tested on data from a real sensor network application: monitoring the temperatures of the disk drive racks at a data center. Simulations show that in terms of estimation accuracy and stability, the proposed approach outperforms existing average-based interpolation methods, and is more robust to burst missing along the time dimension.

In this paper we present some channel measurement results of distributed multiple-input multiple-output (D-MIMO) systems inside a MD-82 aircraft. Channel capacity and spatial correlation are analyzed from the measured channel matrix. Capacity results of 7 different systems (six D-MIMO systems and a SISO system) are discussed. It is found that D-MIMO systems yield significant capacity gains as compared to SISO system. Spatial correlations among transmitter (Tx) antennas distributed above the cabin aisle are statistically analyzed from the measured data. Result shows that, when receiver (Rx) antenna array is placed in the front or back of the cabin, some Tx antenna pairs are highly correlative (correlation coefficients are higher than 0.8). The effect of Tx antenna selection on capacity characteristic of in-cabin D-MIMO system is also studied. It is found that Tx antenna selection can further improve channel capacity and the optimum selected Tx antennas are usually the ones near to the Rx array.

Currently, the mediastreamer2 contains internal support for G.711u, G.711a, speex and gsm. However, it does not contain internal support for the popular G.729, which is known as the best audio codec ever. In this paper, we present the method of how to implement a G.729 codec filter on mediastreamer2. Then, we evaluate the performance of the G.729 filter in our platform. The experimental results show that our implementation of G.729 codec on mediastreamer2 is successful on the whole.

Abstract IEEE 802.11 protocol supports adaptive rate mechanism, which selects the transmission rate according to the condition of the wireless channel, to enhance the system performance. Thus, research of multi‐rate IEEE 802.11 medium access control (MAC) performance has become one of the hot research topics. In this paper, we study the performance of multi‐rate IEEE 802.11 MAC over a Gaussian channel. An accurate analytical model is presented to compute the system saturation throughput. We validate our model in both single‐rate and multi‐rate networks through various simulations. The results show that our model is accurate and channel error has a significant impact on system performance. In addition, our numerical results show that the performance of single‐rate IEEE 802.11 DCF with basic access method is better than that with RTS/CTS mechanism in a high‐rate and high‐load network and vice versa. In a multi‐rate network, the performance of IEEE 802.11 DCF with RTS/CTS mechanism is better than that with basic access method in a congested and error‐prone wireless environment. Copyright © 2008 John Wiley & Sons, Ltd.