Due to data heterogeneity and parameter transmission failure, federated learning (FL) usually suffers performance degradation in wireless networks. To address this issue, a novel federated Learning via reusing historical Information (FedRe) is proposed for convergence acceleration and performance enhancement without any additional communication cost. Specifically, an analytical model is first built to characterize the impacts of data heterogeneity and parameter transmission failure on FL performance. Based on this, a local gradient correction method and a statistical aggregation correction method are proposed to deal with data heterogeneity and transmission unreliability problems respectively. Additionally, the convergence of the proposed FedRe is proved and analyzed theoretically. Finally, extensive numerical results are presented with experiments on two public datasets to validate the effectiveness of FedRe with comparisons of baselines.

Low Earth Orbit (LEO) satellite communications have emerged as a promising paradigm for achieving ubiquitous coverage, driving the evolution of space-ground integrated networks (SGINs). The cell-free (CF) architecture has attracted significant attention in SGINs as the terrestrial segment for its potential to enhance capacity and connectivity. However, deploying CF necessitates numerous access points (APs), resulting in a prohibitive cost. To this end, we propose reconfigurable intelligent surface (RIS)- and simultaneous transmitting and reflecting (STAR)-RIS-assisted CF systems for SGINs, where part of the APs is replaced with cost-efficient RISs and STAR-RISs. Non-orthogonal multiple access (NOMA) is incorporated to improve connectivity under limited spectrum. We formulate transmit power minimization problems for both RIS- and STAR-RIS-assisted CF-NOMA in SGINs, jointly optimizing the active beamforming vectors of the satellite and APs, as well as the discrete passive beamforming (DPB) vectors of RISs/STAR-RISs. For the RIS-assisted scenario, a semi-definite programming (SDP)-based method is proposed to optimize the active beamforming vectors, while an enhanced integer linear programming (ILP) method is proposed to obtain the optimal DPB of RISs. To reduce complexity, we develop a low-complexity penalty-based SDP (PB-SDP) algorithm that achieves near-optimal DPB solutions. For the STAR-RIS-assisted scheme, both independent and coupled DPB for transmission and reflection are optimized alone with the active beamforming vectors. Numerical results demonstrate that: 1) The proposed systems outperform cell-based systems and heuristic optimization algorithms in terms of transmit power consumption; 2) The proposed PB-SDP algorithm achieves near-optimal performance with reduced complexity; 3) It is shown that DBP with 3 quantization bits achieves performance comparable to continuous passive beamforming (CPB) in both RIS- and STAR-RIS-assisted systems; 4) Also, it is shown that beyond a certain number of APs, further increasing the APs yields only limited transmit power consumption gains under a fixed total number of antennas.

Specific Emitter Identification (SEI), based on the inevitable hardware imperfections of emitters, plays a crucial role in physical layer security, independent of upper-layer cryptographic methods. In recent years, SEI combined with deep learning has demonstrated exceptional performance and considerable potential. However, most SEI methods are designed for closed, independent identically distributed environments, which contrasts with real-world open scenarios. This discrepancy leads to significant security vulnerabilities when deploying models in practical applications. To address this challenge, we propose a novel Open-Set SEI (OS-SEI) framework, Outlier Exposure with Label Smoothing (OELS), which utilizes partial exposure to auxiliary anomalous data during the training phase and introduces specific loss functions to mitigate the open-set risk. By constructing smooth decision boundaries through label smoothing and applying thresholds to class confidence, our SEI framework demonstrates robust open-set authentication capabilities. Extensive experiments conducted on ADS-B and WiFi datasets show that the OELS method achieves outstanding open-set recognition performance, with F1 scores of 83.77% and 93.14% at an openness level of 0.10557. Furthermore, ablation experiments confirm the effectiveness and potential of our framework’s components. By increasing the diversity of auxiliary unknown data, we further improve performance in OS-SEI tasks.

Specific emitter identification (SEI) is crucial in the Internet of Everything (IoE). Over the past decade, deep learning (DL) and broad learning (BL)-enabled SEI technologies have emerged. Both DL- and BL-based SEI methods rely on extensive radio frequency (RF) signal samples and corresponding labels, but labeling unknown signals is a considerable overhead and costly task. Consequently, many researchers have begun exploring semi-supervised learning techniques to address the semi-supervised SEI (SS-SEI) problem with limited labeled RF signals. However, existing SS-SEI solutions often prioritize identification performance, leading to high computational overheads and lacking iterability and scalability. To overcome these challenges, this paper proposes a novel SS-SEI solution, termed deep cloud and broad edge (DCBE). This approach integrates a DL-based SEI method at the cloud server with an updatable BL-based SEI method at the edge node. Initially, several DL-based SEI models are trained using labeled historical data at the cloud server. Meanwhile, an updatable BL-based SEI method is deployed locally on the edge node to identify unlabelled signals. When the DCBE solution is operational, edge nodes capture real-time unlabelled RF signals. The pre-trained DL-based SEI method and the locally BL-based SEI method jointly identify these RF signals. The identification results, along with the new real-time RF signals, are then used to update the weights of the BL-based SEI method at the edge nodes. The DCBE SS-SEI solution is validated using an open-source, large-scale, real-world automatic dependent surveillance-broadcast (ADS-B) dataset. Experimental results demonstrate that the proposed DCBE solution offers significant advantages in terms of SS-SEI performance, reduced computational overhead without GPU dependency, and system robustness in complex environments.

As Industry 4.0 continues to evolve, the integration of cyber-physical systems (CPS) into modern engineering systems marks a significant paradigm shift that not only enhances operational efficiency but also opens up new possibilities for innovation and improved quality of life across various sectors. In intricate communication environments, the precise identification of device identities within CPS holds significant importance for augmenting reliability and robustness. For this purpose, we introduce a robust method for specific emitter identification that utilizes a deep residual shrinkage network, aimed at enhancing the model's ability to accurately recognize emitters, even when operating under conditions of low signal-to-noise ratios. This is an end-to-end recognition method that reduces the dependence on expert knowledge. Through the utilization of specially crafted subnetworks, adaptive thresholding is employed to enable each in-phase and quadrature (IQ) signal to possess its unique set of thresholds. The proposed approach reduces noise influence on the model by incorporating a soft threshold within the nonlinear transformation layer of the deep architecture. Experimental results using real-world data demonstrate that this method surpasses the performance of commonly utilized state-of-the-art specific emitter identification models.

Deep learning-based recognition of Unmanned Aerial Vehicles (UAVs) has become a critical tool for enhancing UAV control through improved accuracy and efficiency. However, the practical deployment of these systems is often hampered by the costly acquisition and scarcity of annotated data, which challenges the generalizability of the models. To address this bottleneck, our study employs semi-supervised (SS) learning strategies to exploit the untapped potential of unlabeled data effectively. We introduce a novel semi-supervised approach for UAV recognition that utilizes a self-adaptive threshold mechanism. This technique features Self-adaptive Threshold (SAT) and Self-adaptive Fairness (SAF) mechanisms, designed to dynamically optimize threshold values and guarantee a balanced distribution of labels among various classes. Our method is rigorously evaluated against a comprehensive, open-source UAV dataset. The findings indicate that our semi-supervised model significantly outperforms existing supervised learning models, static threshold SS approaches, and generative models, especially in scenarios with a limited amount of labeled data. These results underscore the effectiveness of our approach in enhancing the practicality and applicability of UAV recognition systems.

Automatic modulation classification (AMC) plays an important role in the field of physical layer security, providing a new way to enhance the security of data transmission and anti-interference ability. Recently, deep learning (DL) has been widely applied in radar and communication signal classification, which requires sufficient labeled training samples to achieve high classification accuracy. However, in non-cooperative situations, it is difficult to obtain a large number of labeled signal samples. Therefore, we propose a novel few-shot AMC method using architecture search and knowledge transfer. This method first utilizes the state-of-the-art neural architecture search algorithm, A-DARTS, to automatically search for the optimal network structure (i.e., Auto-MCNet) based on the auxiliary sample set. Then, the Auto-MCNet model is pre-trained on the auxiliary dataset to explore prior knowledge about signal classification. Finally, we transfer this knowledge to a few-shot training dataset and fine-tune the Auto-MCNet model to enhance its generalization ability. The simulation results show that compared to advanced competitors, Auto-MCNet achieves higher classification accuracy with lower model complexity.

Deep learning (DL), renowned for its superior feature extraction capabilities, has remarkably succeeded in specific emitter identification (SEI), especially when supported by high-quality labeled data. However, obtaining accurate signal labels in complex electromagnetic environments is challenging, and manual labeling is prone to errors, underscoring the need for robust DL-based SEI methods that can handle label noise. These methods prevent neural networks from overfitting noisy labels, thereby boosting identification performance. Yet, research in this area is still limited. Our study introduces a robust label-noise SEI approach and the sample selection and regularization (SSR) method. This involves a two-stage adaptive sample selection (ASS) driven by confidence learning. The first stage entails coarse-grained separation of true and false labels through direct deep neural network (DNN) training. In the second stage, semi-supervised learning (SSL) utilizes a regularization-inspired loss, incorporating label smoothing regularization (LSR) and entropy minimization (EM), for fine-grained sample selection. The DNN is ultimately trained on precisely selected true-labeled samples. Comparative experiments on the automatic dependent surveillance-broadcast (ADS-B) and Wi-Fi data sets demonstrate that our SSR method outperforms the existing methods in identification accuracy, particularly at a 20% label-noise ratio, achieving 86.00% accuracy with the ADS-B data set, and 99.38% with the Wi-Fi data set. The code is available at:https://github.com/sleepeach/SSR-SEI.

Deep learning techniques were widely adopted in various scenarios as a service. However, they are found naturally exposed to adversarial attacks. Such imperceptible-perturbation-based attacks can cause severe damage in nowaday authentication systems that adopt DNNs as the core, such as fingerprint liveness detection systems, face recognition systems, etc. This paper avoids improving the model's robustness and realizes the defense against adversarial attacks based on denoising and reconstruction. Our proposed method can be viewed as a two-step defense framework. The first step denoises the input adversarial example, then reconstructing the sample to close to the original clean image and help the target model output the original label. The proposed method is evaluated using six kinds of state-of-art adversarial attacks, including the adaptive attacks, which are known as the strongest attacks.We also specifically focus on demonstrating the effectiveness of our proposed work in Finance Authentication systems as a real-life case study. Experimental results reveal that our method is more robust than the previous super-resolution-only defense in respect of attaining a higher averaging accuracy over clean and distorted samples. To the best of our knowledge, it's the first work that reveals a comprehensive defense framework against adversarial attacks over Finance Authentication systems.

Radio frequency fingerprinting (RFF) serves as a distinctive hardware trait in transmitters, forming the cornerstone of transmitter identification. While recent advancements led to significant improvements in identification accuracy, these developments also inadvertently simplify the process for adversaries to detect our transmitters. This vulnerability is particularly concerning in secure communications, as the exposure of device information could potentially result in the compromise of communication content, posing significant security threats. To counteract such risks and safeguard transmitters against unauthorized identification, this paper proposes a novel RFF concealment (RFFC) method based on IQ imbalance compensation and digital pre-distortion (DPD) techniques. This method not only effectively conceals the RFF, preventing malicious detection of the transmitter, but also enhances the system’s linearization performance. The effectiveness of the proposed RFFC framework is validated through MATLAB Simulink and a software and hardware test platform. Experimental results show that using the blind generalized linear structure-based IQ imbalance and deep neural network (DNN)-based PA nonlinearity joint concealment method performs best, reducing transmitter identification accuracy to only 17% under various signal-to-noise ratio conditions. Additionally, this method performs the best in system linearization performance.

The advent of 5G opens up hitherto unimagined opportunities for delivering the much-anticipated Internet of Vehicles (IoV). There are two typical types of services in IoV, i.e., safety service and non-safety service. However, the existing IoV framework cannot efficiently support the differentiated IoV services. In addition, maximizing the number of accessed vehicles/users is another critical issue in IoV, especially in the dense urban area. To address these problems, in this paper, we utilize the emerging network slicing technology to support different services and employ Non-Orthogonal Multiple Access technology (NOMA) to help increase the connectivity. In particular, for the safety service, we take advantage of the finite blocklength capacity to correctly record the delay. Our goal is to maximize the connectivity of users by jointly considering the user association and their beamforming vectors, under the restrictions of limited physical resource. The problem is formulated as a Mixed-Integer Nonlinear Programming problem (MINLP). To tackle the intractable MINLP, we propose a two-stage scheme. Firstly, we exploit efficient approaches to solve the beamforming problem, i.e., successive convex approximation, semidefinite relaxation and second-order cone programming. Secondly, we propose a low-complexity Greedy User Association (GUA) algorithm to solve the user association problem. Finally, comprehensive simulations verify that our proposed GUA algorithm is close to global optimal solution and outperforms the benchmark schemes.

The rapid development of an unmanned aerial vehicle (UAV) has brought new opportunities for wireless communication and edge computing. In this article, we investigate the scenario where multiple UAVs serve as edge computing devices for the Internet of Vehicles (IoV). Regardless of the allocation of computing resources, we focus on bandwidth allocation and trajectory control to maximize the communication capacity of the system so that the UAV edge computing network can process more data. With this intent, a UAV-assisted IoV edge computing system model is constructed as a nonconvex optimization problem, aiming to maximize the achievable channel capacity of the network. To solve this problem, two “quasi-distributed” multiagent algorithms, i.e., actor-critic mixing network (AC-Mix) and multi-attentive agent deep deterministic policy gradient (MA2DDPG), are proposed based on deep deterministic policy gradient. Specifically, AC-Mix utilizes a mixing network to obtain a global$Q$-value for better evaluation of joint action, while MA2DDPG employs a multihead attention mechanism to achieve multiagent collaboration. Using multi-agents deep deterministic policy gradient (MADDPG) as benchmark, several experiments are carried out to verify the performance of the proposed algorithms. Simulation results show that the convergence velocity of AC-Mix and MA2DDPG is improved by 30.0% and 63.3%, respectively, compared with MADDPG.

Accurate crowd flow prediction has gained increasing importance for the development of social Internet-of-Things (IoT) systems. In this article, we provide an efficient crowd flow prediction for social IoT systems in urban space based on the mobile network big data. In particular, the usage detail records (UDRs) are used in the prediction. The feasibility of using UDRs in the prediction is first analyzed. Then, a graph data model is exploited to record and represent the mobile behavior of users. In particular, we propose to apply the heterogeneous information network (HIN) representing the UDR data and characterize the users’ behavior through the embedding methods of HIN. Moreover, an attention-based spatiotemporal graph convolution network with embedded vectors (EA-STGCN) is proposed for the final prediction. Through experimental evaluation, the advantages of the proposed model are shown in comparison to benchmarks.

This paper studies an unmanned aerial vehicle (UAV)-enabled wireless power communication networks (WPCN-s), where the UAV provides energy for mobile user nodes (M-UNs) and receives information from M-UNs. The movement of M-UN complies with a Gauss-Markov random model. To ensure acceptable quality-of-service (QoS), we consider dynamically planning the flight path of the UAV according to the movements of M-UNs. Since the flight time of UAV is restricted by limited energy, nonorthogonal multiple access (NOMA) is adopted to access a large number of M-UNs for simultaneous information transmission. Based on the above considerations, we aim to maximize the throughput via path planning of the UAV, subject to the QoS requirements of M-UNs and the UAV's energy constraint. To handle the challenges brought by dynamically changing channels to solving the problem, we propose a QoS-based double deep Q-learning network (DDQN). Numerical simulation results show that, compared with the conventional algorithms, the proposed framework achieves higher throughput.

Mobile edge computing is a novel paradigm that provides computing capabilities at the edge of the radio access network close to end devices to support latency-critical applications and services. However, the benefits will be canceled out by the limited computing capacity of edge servers. An edge-cloud paradigm has been studied to improve computing capabilities to solve the above problem. In this paper, a multi-cell edge-cloud architecture is considered to meet the demands of latency-sensitive applications and deal with large-scale data offloading. The optimized offloading scheme is studied to minimize the devices' overhead which is measured as a function of energy consumption and computational cost. We formulate the problem as a Mixed Integer Linear Programming Problem and adopt the Branch-and-Bound algorithm to solve it. Due to the high time overhead of the method, we first transform the problem into a more tractable form and then adopt a learning approach to imitate the branching strategy to improve the Branch-and-Bound algorithm. Experiments of results show that our approach can reduce the time-cost of the Branch-and-Bound and the result is close to the traditional scheme.