Unmanned aerial vehicles (UAVs) in wireless communication systems offer rapid deployment, flexible reconfiguration, and superior communication channels, thanks to their short-range line-of-sight links, making them more efficient and cost-effective than terrestrial or high-altitude platform networks. As a promising technique, integrated sensing and communication (ISAC) enhances UAV networks by integrating sensing and communication functionalities. This concurrent design improves spectrum efficiency and reduces hardware costs. To ensure reliable communication for their high mobility, a novel modulation technique, the orthogonal time-frequency space (OTFS) waveform, has been proposed, which leverages the delay-Doppler domain for efficient information transmission. In this paper, we investigate ISAC-based multi-target sensing with distributed OTFS radars to deliver reliable performance for UAV networks. To achieve that, we propose to leverage the delay and Doppler information featured by OTFS signals to determine the range and radial velocity, accomplishing successful sensing tasks. Moreover, to address the challenge of unassociated measurements and targets, we propose a novel optimization framework to concurrently perform data association and target sensing tasks. This framework is developed by formulating a mixed-integer optimization problem, which is then solved with polynomial complexity through convex approximation. Additionally, we propose an iterative maximum likelihood estimator (MLE) to further enhance sensing performance by accounting for target-measurement errors. Extensive simulation results verify the superiority of our proposed work to state-of-the-art methods.

This paper investigates the receiver design for clipped orthogonal time frequency space (OTFS) systems, where the user devices are equipped with power amplifiers (PAs) with low dynamic range. To improve power efficiency, the PAs have to work near the saturation points, which leads to unknown nonlinear distortions, thus making the signal detection more challenging. To solve this problem, techniques like intentional clipping or pre-distortion are adopted, thus approximating the outputs of the PAs as clipped signals. To further compensate for the unknown time-varying multipath channel and the clipping distortion at the receiver, the channel and clipping amplitude (CA) estimation, channel tracking, and signal detection are studied in this paper. Firstly, a receiver framework is developed for clipped OTFS. Secondly, by adopting the sparsity of the delay-Doppler (DD) domain channel and the piecewise linearized signal model with respect to CA, a novel sparse Bayesian learning (SBL) based joint channel and CA estimation scheme is proposed. Then, to further reduce the estimation error and bit error rate, a Kalman filter (KF) based channel tracking scheme and a minimum mean square error decision feedback blockwise equalization (MMSE-DFBE) based detection scheme are proposed. These two schemes are integrated in an expectation maximization (EM) based iterative tracking and detection algorithm. Finally, numerical simulations are conducted to demonstrate the superiority of the proposed schemes in terms of both estimation error and bit error rate.

With the rapid expansion of Internet of Things (IoT) applications, the demand of wide coverage and massive connectivity is inevitable. In this context, this paper investigates massive unsourced random access (URA) paradigm for low earth orbit (LEO) satellite IoT applications, focusing on device separation and signal detection. By exploiting the structured Grassmannian constellation to generate the codebook, a tensor-based URA transmission scheme is provided, which models the separation and detection problem as a general canonical polyadic (CP) decomposition. Then, to evaluate the access capability of our considered URA scheme, a comprehensive uniqueness analysis considering both sufficient conditions and necessary conditions is presented. Accordingly, an efficient generalized line-search-accelerated alternating least squares (GLSA-ALS) method is proposed to conduct the device separation and signal detection, which can avoid a large number of inverse computations for large-scale matrices. To be specific, with the help of the relaxation factors during the iteration, our proposed method can converge at a fast speed with negligible performance loss, which facilitates a better trade-off between the detection accuracy and computational complexity. Furthermore, depending on the demand of a specific application scenario, the flexible selection of relaxation factors enables the proposed method to be compatible to the classical ALS method, which can enhance the performance at the cost of additional complexity. Finally, relying on the maximum likelihood (ML)-based detection approach, the message list transmitted by active devices from one common codebook can be recovered. Simulation results demonstrate that the proposed GLSA-ALS method outperforms the state-of-the-art methods for practical LEO satellite IoT applications.

Beamforming is a key technology for achieving integrated sensing and communication (ISAC). However, most existing works focus on mono-static sensing, which has limited sensing accuracy and strong self-interference. To address these issues, this paper investigates hybrid beamforming (HBF) design for millimeter-wave (mmWave) multiple-input multiple-output (MIMO) ISAC system with multi-static cooperative localization. Specifically, one access point (AP) simultaneously forms communication beams to serve multiple user equipments (UEs) and a sensing beam towards one target, and other multiple distributed APs perform cooperative localization on the target by estimating the angle-of-arrivals (AOAs) of received echo signals. First, to characterize the target localization accuracy, we derive the squared position error bound (SPEB) of AOA-based multi-static cooperative localization. Then, two HBF optimization problems are formulated to investigate the performance tradeoff between sensing and communication. For the sensing-centric design, we aim to minimize the SPEB of target localization while ensuring the signal-to-interference-plus-noise ratio (SINR) requirements of individual UEs. To tackle this nonconvex problem, we propose a semidefinite relaxation (SDR)-based alternating optimization algorithm. For the communication-centric design, a fractional programming (FP)-based alternating optimization algorithm is proposed for solving the communication sum-rate maximization problem under the sensing SPEB constraint. Simulation results demonstrate that the proposed two HBF algorithms can achieve localization accuracy and sum-rate performance close to fully-digital beamforming counterparts and outperform other baseline schemes.

Long-range (LoRa) can provide highly energy-efficient and cost-effective communications for low power wide area networks, playing an indispensable role in the Internet of Things (IoT). However, terrestrial LoRa networks cannot guarantee pervasive connectivity, especially in rural and remote areas. To tackle this problem, exploiting LoRa-based low Earth orbit (LEO) satellite IoT has garnered a growing interest in both academia and industry. In this paper, we provide a novel analytical framework based on spherical stochastic geometry (SG) for characterizing the uplink access probability of LoRa-based LEO satellite IoT. For practical modeling, multiple classes of LoRa end-devices (EDs) are taken into consideration, where each class of EDs is modeled by an independent Poisson point process (PPP). Both the channel characteristics of near-Earth satellite communications and the unique features of LoRa network are considered to derive closed-form analytical expressions for the uplink access probability. Numerical simulations validate the accuracy of our theoretical analysis and provide insightful guidelines for the practical design and implementation of LoRa-based LEO satellite IoT.

Beamforming is a key technique for achieving integrated sensing and communication (ISAC). However, most existing works focus on mono-static sensing, which can only provide limited sensing accuracy and range. In this paper, we investigate hybrid beamforming design for millimeter-wave (mmWave) multipleinput multiple-output (MIMO) ISAC system with multi-static cooperative localization, where one access point (AP) simultaneously transmits communication beams to serve multiple user equipments (UEs) and transmits a sensing beam towards a target, and other nearby APs perform cooperative localization on the target by estimating the angle-of-arrivals (AOAs) of received echo signals. To characterize the target localization accuracy, we derive the squared position error bound (SPEB) of AOA-based cooperative localization by using the equivalent Fisher information matrix (EFIM). Then, the hybrid beamforming design problem is formulated to minimize the SPEB of target localization, while satisfying the signal-to-interference-plus-noise ratio (SINR) requirements of individual communication UEs, transmit power budget, and constant modulus constraints. To solve the non-convex problem, a semidefinite relaxation (SDR)-based alternating optimization algorithm is proposed. Simulation results demonstrate that the proposed hybrid beamforming can achieve localization accuracy close to fully-digital beamforming and outperform the baseline schemes.

Integrated sensing and communication (ISAC) can realize communication and sensing functionalities simultaneously by sharing spectrum and hardware resources, where the sensing performance can be guaranteed by accurate range and velocity estimation. However joint range and velocity estimation inherently confronts the accuracy-complexity tradeoff. Therefore, a low-complexity joint range and velocity estimation algorithm is developed in this work, referred to as the particle swarm optimization reconstructed subspace multiple signal classification (PSO-RS-MUSIC). The proposed algorithm leverages optimized subspace reuse mechanisms to enhance estimation accuracy. To address the high complexity problem, the PSO-RS-MUSIC algorithm employs the particle swarm optimization (PSO) technique to replace the traditional spectral peak search, thereby reducing computational complexity significantly. Simulation results illustrate that the proposed algorithm outperforms the conventional RS-MUSIC algorithm, while the computational complexity is reduced by more than 90%.

Holographic multiple-input multiple-output (HMIMO) is envisaged as a viable manner for manipulating electromagnetic field produced or perceived by antennas, in an effort to achieve an intelligent and endogenously holography-capable wireless propagation environment. This notion garners significant interest when engaging large antenna elements at high frequencies, such as millimeter-wave or terahertz. Under these conditions, operations often occur within the Fresnel region, i.e., near-field region, where assumptions of planar wavefront no longer apply. This article investigates the codebook solution for HMIMO, unmasking a number of challenges intrinsic in the near-field context and limitations of applying codebooks specified in current standards. Specifically, we proposed a two-phase codebook design empowered by artificial intelligence (AI)-based techniques, where angular and distance ingredients are resolved respectively at each phase, functioning in tandem for facilitating an efficient beam training at reduced pilot overhead. Then from the 3rd generation partnership project (3GPP) standardization perspective, we share potential design rationales influencing standardization, along with a novel signaling procedure for HMIMO beam sweeping.

In this paper, simultaneously transmitting and reflecting reconfigurable intelligent surface (STAR-RIS)-assisted integrated sensing and communication (ISAC) is considered. Based on constraints of communication signal-to-interference-plus-noise ratio (SINR), maximum transmit power limitation, and the law of conservation of energy, the objective is to maximize the SINR of the radar perception signal. To address this non-convex maximization problem, a novel method rooted in fractional programming (FP) and block coordinate descent (BCD) is introduced. To handle non-convex constraints, an alternative optimization algorithm grounded in BCD is introduced. To tackle the non-convex problem of fractional form, the original formulation is optimized utilizing fractional programming techniques, thereby transforming it into a convex problem for more efficient solution. Additionally, select conditions are relaxed through the application of semi-definite relaxation (SDR) techniques. Finally, the numerical results show that: 1) As the number of iterations increases, the proposed algorithm shows good convergence. 2) The performance of the proposed algorithm is significantly better than the state-of-the-art algorithms.

In this paper, we investigate near-field hybrid beam-forming design for millimeter-wave (mmWave) integrated sensing and communication (ISAC) systems, where one base station (BS) equipped with large-scale antenna array simultaneously serves multiple communication users and performs target localization by exploiting the degrees of freedom in both angle and distance domains. First, to characterize the target localization accuracy, we analyze the squared position error bound (SPEB) for estimating the two-dimensional (2D) position of target. Then, the hybrid beamforming design is formulated to maximize the sum-rate of communication users, while guaranteeing the SPEB constraint of target localization, transmit power constraint, and constant modulus constraints. To tackle the nonconvex problem, we propose a fractional programming (FP) and successive convex approximation (SCA)-based block coordinate descent (BCD) algorithm. Simulation results demonstrate that the proposed hybrid beam-forming can achieve sum-rate close to fully-digital beamforming and outperform the baseline schemes.

With the rapid development of Internet of Things (IoT), efficient and reliable massive IoT device connections need to be widely supported in the upcoming next-generation communication networks, especially for emerging high-mobility scenarios. In this context, this paper investigates massive grant-free random access (GF-RA) in high mobility scenarios, focusing on active device detection (ADD) and channel estimation (CE) under fast time-varying channels. By exploiting the inherent low-rank structure of the observed pilot-signal-tensor, a tensor-based GF-RA transmission scheme is provided. On this basis, we propose a joint ADD and CE method based on the canonical polyadic (CP) model for both sourced and unsourced RA frameworks. More specifically, by remodelling the observation signal as a third-order tensor, the channel parameters can be grouped in the factor matrices of the CP model. However, the excessive number of potential device connections in massive GF-RA scenarios lead to excessively large dimensions of the factor matrices, thus resulting in severe ill-condition. To solve this problem, the Vandermonde structure of factor matrices is developed, which enables the effective exploitation of the tensor subspace for CP decomposition. Then, by utilizing the pre-allocated training precoders, an effective two-dimensional search method is proposed to jointly detect active devices and initialize the iterative estimation of channel parameters. Finally, due to the grouping situation, independent and coupled channel parameters are estimated by appropriate methods based on maximum likelihood (ML) and iterative updating, respectively. Moreover, the pre-allocation of training precoders can be unified to the unsourced RA scenarios, where the joint ADD and CE can be regard as a simple degenerate method compared to sourced RA. Simulation results demonstrate that the proposed tensor-based GF-RA framework outperforms the state-of-the-art schemes in terms of both ADD and CE performance.

Recently, Long-Range (LoRa) based low Earth orbit (LEO) satellite Internet of Things (IoT) has garnered growing interest from both academia and industry, since it can guarantee pervasive connectivity in an energy-efficient and cost-effective manner. In this paper, we provide a novel spherical stochastic geometry (SG) based analytical framework for characterizing the uplink access probability of LoRa-based LEO satellite IoT system. Specifically, multiple classes of LoRa end-devices (EDs) are taken into consideration, where each class of LoRa EDs is modeled by an independent Poisson point process (PPP). Both the channel characteristics of the satellite-to-Earth communications and the unique features of the LoRa network are considered to derive closed-form analytical expressions for the uplink access probability of such a new paradigm. Moreover, the non-trivial impact of the spreading factor, the ED’s density, the orbit altitude, and the satellite effective beamwidth on the system performance is thoroughly investigated. Extensive numerical simulations are conducted, which not only validate the accuracy of our theoretical analysis but also provide useful insights into the practical design and implementation of LoRa-based LEO satellite IoT system.

LoRa is one of the most prominent low-power wide area network (LPWAN) technologies for Internet of Things (IoT) applications. As the core technique of LoRa physical (PHY) layer, chirp spread spectrum (CSS) modulation is employed to support low power and long range communications. Although it provides a compelling tradeoff between coverage and data rate, the relatively low-spectral efficiency (SE) is still a limiting factor for its extensive applications. In this article, we propose two enhanced group-based CSS modulation schemes, named in-phase and quadrature group-based CSS (IQ-GCSS) and time domain multiplexed group-based CSS (TDM-GCSS), which can achieve much higher SE than the conventional LoRa modulation. The transmitter architectures of our proposed modulation schemes are presented along with both coherent and noncoherent detection methods. Moreover, an overall performance analysis of our proposed schemes is provided in terms of bit error rate (BER) and computational complexity. Numerical results not only validate the accuracy of our theoretical analysis but also demonstrate substantial performance improvements of our proposed schemes in terms of effective throughput compared to the classical counterparts.

This paper proposes a novel integrated sensing and communication (ISAC) receiver design framework for OTFS (orthogonal time frequency space)-based MIMO (multi-input-multi-output) systems from a unified perspective of variational inference. We first construct a factor graph representation for the OTFS-based MIMO system according to the factorization of the a posteriori probability (APP). This representation establishes a direct probabilistic link between sensing and communication, allowing both functionalities to benefit from their integration. On this basis, we develop a low computational complexity message passing algorithm by minimizing the variational free energy associated with the global APP. In particular, belief propagation, mean field, and expectation maximization algorithms for data detection, channel coefficient estimation, and kinematic parameter sensing are derived, respectively. To reduce the communication overhead for the implementation of ISAC algorithm, we propose a federated learning scheme for distributed kinematic parameter sensing. Specifically, by solving the sensing problem in different fashions, three federated learning modes are devised. Simulation results validate the superior performance of the proposed scheme.

In recent years, long-range (LoRa) has become one of the most prominent low-power wide-area network (LPWAN) technologies for the Internet of Things (IoT), which is based on a proprietary chirp spread spectrum (CSS) modulation (i.e., LoRa modulation). However, with the ever-increasing transmission demands of various IoT applications, the low-data-rate issue of LoRa modulation has become a critical bottleneck for its extensive deployment. To address this issue, we first formulate a unified framework for CSS-based waveform design and propose a novel layered group-based CSS (LGCSS) modulation scheme to achieve much higher spectral efficiency (SE) and data rate, thus accommodating a wider range of IoT applications. The complete transmitter architecture of LGCSS modulation is presented along with both coherent and non-coherent detection methods. Moreover, a comprehensive performance analysis of our proposed scheme is conducted in terms of orthogonality, bit error probability (BEP), and computational complexity. Extensive numerical simulations are conducted to verify the effectiveness of our theoretical analysis and the superiority of our proposed scheme compared to the traditional counterparts.

In this paper, we develop a graph neural network (GNN)-assisted bilinear inference approach to enhance the receiver performance of the MIMO system through message passing-based joint channel estimation and data detection (JCD). Specifically, based on the bilinear generalized approximate message passing (BiG-AMP) framework and conditional correlation of signal, we propose a GNN-assisted BiG-AMP (GNN-BiGAMP) approach, which integrates a GNN module into the data-detection-loop to compensate the inaccurate marginal likelihood approximation. By leveraging the coupling between the channel and received symbols, a bilinear GNN-assisted BiG-AMP (BiGNN-BiGAMP) JCD receiver is further proposed. This method incorporates two GNNs with similar graph representation into the bilinear posterior estimation loops, which not only compensates for approximation errors but also alleviates performance loss due to premature variance convergence, thereby enhancing the receiver performance significantly. To fully exploit the supervised information from channel estimation and data detection, we propose a multitask learning based training scheme, which coordinates GNNs with different tasks in two loops. Simulation results show that our proposed GNN-assisted JCD receivers significantly outperform other JCD counterparts in terms of both channel estimation and data detection.

Action recognition algorithms have gained significant attention in recent years, which can be indispensable for a plethora of cutting-edge applications like extended reality or Metaverse. These services often pose stringent requirement on immediate sensing and cognition of the surroundings, which necessitates immediate classifications of the captured actions (e.g., video data) that classical signal processing methods can hardly attain. In this paper, we introduced a residual artificial neural network with two-stream structure to further improve the accuracy of action recognition algorithm. Specifically, two residual networks (ResNet101) are trained separately, one by spatial RGB image streams, and another by optical flow streams. The two-strem network outputs are then fed into a fusion classifier, in which information extracted by spatial network and temporal network jointly determines the classification result. Moreover, in the training process, hyper-parameters setting and optimizer selection are performed numerically to achieve optimal performance. Finally, the recognition accuracy of the proposed algorithm has been compared to other existing widely-employed counterparts, where UCF101 data set is utilized for training and testing. Simulations validates aiming that the network can achieve higher recognition accuracy than traditional algorithms, and the two-stream method shows its superiority over the single-network counterpart.

Abstract Within the evolving landscape of fifth‐generation (5G) wireless networks, the introduction of network‐slicing protocols has become pivotal, enabling the accommodation of diverse application needs while fortifying defences against potential security breaches. This study endeavours to construct a comprehensive network‐slicing model integrated with an attack detection system within the 5G framework. Leveraging software‐defined networking (SDN) along with deep learning techniques, this approach seeks to fortify security measures while optimizing network performance. This undertaking introduces network slicing predicated on SDN with the OpenFlow protocol and Ryu control technology, complemented by a neural network model for attack detection using deep learning methodologies. Additionally, the proposed convolutional neural networks‐long short‐term memory approach demonstrates superiority over conventional ML algorithms, signifying its potential for real‐time attack detection. Evaluation of the proposed system using a 5G dataset showcases an impressive accuracy of 99%, surpassing previous studies, and affirming the efficacy of the approach. Moreover, network slicing significantly enhances quality of service by segmenting services based on bandwidth. Future research will concentrate on real‐world implementation, encompassing diverse dataset evaluations, and assessing the model's adaptability across varied scenarios.

Unmanned aerial vehicle (UAV) equipped with large-scale antenna array constitutes a promising relaying candidate for reliable and secure satellite-terrestrial communication. Due to the limitation of energy consumption, a novel UAV architecture with one-bit analog-to-digital converters (ADCs) and one-bit digital-to-analog converters (DACs) is proposed firstly. Leveraging the additive quantization noise model, the exact closed-form expressions of both ergodic capacity and ergodic achievable secrecy rate are derived for UAV-assisted satellite-terrestrial communication systems using large-scale antenna array with one-bit ADCs/DACs. To enhance the transmission capacity and combat the eavesdropper simultaneously, maximum-ratio combining (MRC) is used by UAV to receive signals from satellite and location-based beamforming (LBB) is adopted by UAV to forward signals to destination, where the beamformer is optimized based on the derived expression of the ergodic achievable secrecy rate. Simulation results validate the accuracy of our analytical ergodic achievable secrecy rate, and demonstrate that our proposed MRC/LBB scheme has better secrecy rate than its conventional location-based counterpart.

Hybrid precoding has been considered as a promising technology for millimeter-wave (mmWave) massive multiple-input multiple-output (MIMO) systems, since it can achieve a tradeoff between system performance and hardware complexity. However, the optimal solution is difficult to obtain due to the coupling between analog precoder and digital precoder, as well as the non-convex constant modulus constraint, especially in multiuser scenarios. In this paper, we investigate several hybrid transceiver designs in wideband mmWave massive multiuser MIMO-OFDM systems for maximizing the spectral efficiency. Firstly, we propose two joint designs of hybrid precoder and combiner based on alternating optimization. Specifically, the intractable spectral efficiency maximization problem is reformulated as an equivalent weighted minimum mean square error (WMMSE) problem. To design the analog precoder and combiner with non-convex constant modulus constraint, we develop two efficient algorithms based on majorization minimization (MM) and element-wise block coordinate descent (EBCD) techniques, respectively. Secondly, to reduce the computational complexity, we propose a discrete Fourier transform (DFT) codebook based scheme, which can enhance the beamforming gain and mitigate the inter-beam interference. Thirdly, the convergence and complexity analysis are presented. The proposed two alternating optimization algorithms are guaranteed to converge to locally optimal solutions. Simulation results demonstrate that the proposed hybrid transceiver designs achieve significant performance gains over state-of-the-art schemes.

Huge overhead of beam training poses a significant challenge to mmWave communications. To address this issue, beam tracking has been widely investigated whereas existing methods are hard to handle serious multipath interference and non-stationary scenarios. Inspired by the spatial similarity between low-frequency and mmWave channels in non-standalone architectures, this paper proposes to utilize prior low-frequency information to predict the optimal mmWave beam, where deep learning is adopted to enhance the prediction accuracy. Specifically, periodically estimated low-frequency channel state information (CSI) is applied to track the movement of user equipment, and timing offset indicator is proposed to indicate the instant of mmWave beam training relative to low-frequency CSI estimation. Meanwhile, long-short term memory networks based dedicated models are designed to implement the prediction. Simulation results show that our proposed scheme can achieve higher beamforming gain than the conventional methods while requiring little overhead of mmWave beam training.

In this paper, we investigate secure relay selection for finite-state Markov channel and propose a Q-learning assisted relay selection scheme. Specifically, we firstly analyze the achievable effective secrecy throughput of random selection scheme and optimal selection scheme, respectively, showing that the secrecy performance is highly determined by relay selection methodology. Then, we leverage the Q-learning to learn how to select relay for finite-state Markov channel, which is capable of selecting proper relay with outdated channel state information. Numerical results demonstrate that our proposed Q-learning assisted relay selection scheme can achieve a significant improvement of effective secrecy throughput even with outdated channel information.

TeraHertz (THz) wireless communication constitutes a promising technique of satisfying the ever-increasing appetite for high-rate services. However, the ultra-wide bandwidth of THz communications requires high-speed, high-resolution analog-to-digital converters, which are hard to implement due to their high complexity and power consumption. In this paper, a deep learning-assisted THz receiver is designed, which relies on single-bit quantization. Specifically, the imperfections of THz devices, including their in-phase/quadrature-phase imbalance, phase noise and nonlinearity are investigated. The deflection ratio of the maximum-likelihood detector used by our single-bit-quantization THz receiver is derived, which reveals the effect of phase offset on the demodulation performance, guiding the architecture design of our proposed receiver. To combat the performance loss caused by the above-mentioned distortions, a twin-phase training strategy and a neural network based demodulator are proposed, where the phase offset of the received signal is compensated before sampling. Our simulation results demonstrate that the proposed deep learning-assisted receiver is capable of achieving a satisfactory bit error rate performance, despite the grave distortions encountered.

Huge overhead of beam training imposes a significant challenge in millimeter-wave (mmWave) wireless communications. To address this issue, in this paper, we propose a wide beam based training approach to calibrate the narrow beam direction according to the channel power leakage. To handle the complex nonlinear properties of the channel power leakage, deep learning is utilized to predict the optimal narrow beam directly. Specifically, three deep learning assisted calibrated beam training schemes are proposed. The first scheme adopts convolution neural network to implement the prediction based on the instantaneous received signals of wide beam training. We also perform the additional narrow beam training based on the predicted probabilities for further beam direction calibrations. However, the first scheme only depends on one wide beam training, which lacks the robustness to noise. To tackle this problem, the second scheme adopts long-short term memory (LSTM) network for tracking the movement of users and calibrating the beam direction according to the received signals of prior beam training, in order to enhance the robustness to noise. To further reduce the overhead of wide beam training, our third scheme, an adaptive beam training strategy, selects partial wide beams to be trained based on the prior received signals. Two criteria, namely, optimal neighboring criterion and maximum probability criterion, are designed for the selection. Furthermore, to handle mobile scenarios, auxiliary LSTM is introduced to calibrate the directions of the selected wide beams more precisely. Simulation results demonstrate that our proposed schemes achieve significantly higher beamforming gain with smaller beam training overhead compared with the conventional and existing deep-learning based counterparts.

In this study, the authors investigate the secure transmission of a cooperative network, in which a source communicates with a destination via multiple cooperative nodes in the presence of a full‐duplex active eavesdropper, which can intercept the confidential signals and transmit jamming signals simultaneously. To safeguard the security of legitimate communication, two joint relay and jammer selection schemes are proposed according to the availability of the eavesdropper's channel state information, namely, optimal relay and random jammer selection scheme and optimal relay and optimal jammer selection scheme. The authors first derive the exact closed‐form expressions of the secrecy outage probability (SOP) for different selection schemes. Aiming at minimising SOP, they then adopt the deep feedforward neural network to determine the optimal power allocation between the selected relay and jammer. Further, the asymptotic expressions for SOP in the high signal‐to‐noise ratio regime are derived. Numerical results verify the analysis and demonstrate the performance advantage of the proposed scheme over conventional relay selection scheme with optimal power allocation.

In this work, we propose a secure communication scheme, where a source transmits information to the legitimate receiver in the presence of multiple eavesdroppers. To improve the security, single or multiple friendly jammers are deployed to confuse the eavesdroppers. Specifically, we assume that the jammers have to harvest energy from the source, thus we consider a two-phase transmission scheme where the source transmits energy to the jammers first and then transmits information to the legitimate receiver confidentially with the help of the jammers. We first give the expression of the secrecy outage probability, revealing how the secrecy performance depends on the transmission parameter tuple, and then we use the simulated annealing method to obtain the optimal transmission parameter tuple. The simulation results show the superiority of our proposed scheme.

In this paper, we investigate the physical layer security of a dual-hop cooperative network in the presence of a full-duplex active eavesdropper, which can overhear the confidential signals and transmit jamming signals simultaneously. We utilize the optimal relay selection scheme to improve the secrecy performance, where the relay maximizing the secrecy capacity will be selected to forward the information. To evaluate the secrecy performance of our system, we derive a compact closed-form expression of the secrecy outage probability. Besides, we also analyze the asymptotic performance related to the position of the nodes. Finally, we verify our analysis through the numerical results, and demonstrate that there exists a secrecy protection region where the secrecy outage probability is below a target probability.

In this study, the authors propose a learning‐based approach to improve the security of the authors' considered communication system in a dynamic environment, where a source transmits information to a legitimate receiver in the presence of an active eavesdropper. Additionally, they assume that the source has to harvest energy from the environment to support its communication. Due to the dynamic of the environment, both the harvested energy and the channel vary over time, requiring a dynamic transmission strategy that follows the changes. In order to improve the security performance, they first analyse how to select the optimal transmission parameters in hindsight, and then they propose to combine the Q‐learning algorithm and the expert advice method to maximise the cumulative reward in the dynamic environment. They also introduce an improved learning‐based approach, which accelerates the convergence of their approach. The simulation results show that their proposed learning‐based approach helps the legitimate nodes learn a beneficial transmission strategy to obtain a larger cumulative reward.

Baseband predistortion is a popular and efficient method to linearize high power amplifier (HPA) in wireless communication systems. Polynomial (POLY) and look-up-table (LUT) are two methods to design baseband predistorter (PD). However, on the one hand, POLY-based method is complex to implement. On the other hand, LUT-based predistorter suffers convergence time and quantization error problem. In this paper, we propose a hybrid POLY and LUT predistorter for memory nonlinear system in wideband scenarios, it is also suitable for memoryless channel. Simulations show that the proposed hybrid structure outperforms the traditional one with lower complexity.