In this paper, we propose a new adversarial training framework to address high-dimensional instantaneous channel estimation in wireless communications. Specifically, we train a generative adversarial network to predict a channel realization in the time-frequency-space domain, in which the generator exploits the third-order moment of the input in its loss function and applies a new reparameterization method for latent distribution learning to minimize the Wasserstein distance between the true and estimated channel distributions. Next, we propose an explainable artificial intelligence mechanism to examine how the critic discriminates the generated channel. We demonstrate that our proposed framework is superior to existing methods in terms of minimizing estimation errors. Additionally, we find that the critic's attention focuses on the high-power portion of the channel's time-frequency representation.

Accurate channel estimation is crucial for the improvement of signal processing performance in wireless communications. However, traditional model-based methods frequently experience difficulties in dynamic environments. Similarly, alternative machine-learning approaches typically lack generalization across different datasets due to variations in channel characteristics. To address this issue, in this study, we propose a novel domain adaptation approach to bridge the gap between the quasi-static channel model (QSCM) and the map-based channel model (MBCM). Specifically, we first proposed a channel estimation pipeline that takes into account realistic channel simulation to train our foundation model. Then, we proposed domain adaptation methods to address the estimation problem. Using simulation-based training to reduce data requirements for effective application in practical wireless environments, we find that the proposed strategy enables robust model performance, even with limited true channel information.

This paper studies the statistical characterization of ground-to-air (G2A) and reconfigurable intelligent surface (RIS)-assisted air-to-ground (A2G) communications in RIS-assisted AAV networks under the impact of channel aging. A comprehensive channel model is presented, which incorporates the time-varying fading, three-dimensional (3D) mobility, Doppler shifts, and the effects of channel aging on array antenna structures. We provide analytical expressions for the G2A signal-to-noise ratio (SNR) probability density function (PDF) and the corresponding cumulative distribution function (CDF), demonstrating that the G2A SNR follows a mixture of noncentral$\chi ^{2}$distributions. The A2G communication is characterized under RIS arbitrary phase-shift configurations, showing that the A2G SNR can be represented as the product of two correlated noncentral$\chi ^{2}$random variables (RVs). Additionally, we present the PDF and the CDF of the product of two independently distributed noncentral$\chi ^{2}$RVs, which accurately characterize the A2G SNR’s distribution. Our paper confirms the effectiveness of RIS-assisted communication in mitigating channel aging effects within the coherence time. Finally, we propose an adaptive spectral efficiency method that ensures consistent system performance and satisfactory outage levels when the AAV and the ground user equipment are in motion.

In quasi-static wireless networks characterized by infrequent changes in the transmission schedules of user equipment (UE), malicious jammers can easily deteriorate network performance. Accordingly, a key challenge in these networks is managing channel access amidst jammers and under dynamic channel conditions. In this context, we propose a robust learning-based mechanism for channel access in multi-cell quasi-static networks under jamming. The network comprises multiple legitimate UEs, including predefined UEs (pUEs) with stochastic predefined schedules and an intelligent UE (iUE) with an undefined transmission schedule, all transmitting over a shared, time-varying uplink channel. Jammers transmit unwanted packets to disturb the pUEs’ and the iUE’s communication. The iUE’s learning process is based on the deep reinforcement learning (DRL) framework, utilizing a residual network (ResNet)-based deep Q-Network (DQN). To coexist in the network and maximize the network’s sum cross-layer achievable rate (SCLAR), the iUE must learn the unknown network dynamics while concurrently adapting to dynamic channel conditions. Our simulation results reveal that, with properly defined state space, action space, and rewards in DRL, the iUE can effectively coexist in the network, maximizing channel utilization and the network’s SCLAR by judiciously selecting transmission time slots and thus avoiding collisions and jamming.

In the field of Integrated Sensing and Commu-nication (ISAC) systems, several challenges emerge, such as obtaining the infinitesimal Cramer-Ran lower bound (CRLB) for sensing outcomes and addressing the vulnerabilities of ISAC to adversarial machine learning (AML) attacks. To address this, we propose a Smart ISAC (S-ISAC) system, which incorporates a unique generative adversarial network (GAN) combined with a differentiable Kolmogorov-Smirnov (KS) loss function, named KSGAN. This KSGAN is tailor-made to identify AML attacks on range-Doppler heatmap features. Only after ensuring that the range-Doppler heatmap is free from AML attacks using KSGAN, do we apply the Constant False Alarm Rate (CFAR) for accurate estimation of target vehicle parameters. We implement a rigorous ISAC system under AML attacks using Matlab Toolboxes and the adversarial robustness toolbox (ART). Our numerical findings indicate that the proposed KSGAN offers greater accuracy in detecting AML than a standalone GAN. Additionally, our results show that the MIMO S-ISAC Beamforming surpasses the performance of the standalone ISAC system.

This paper studies the statistical characterization of ground-to-air (G2A) and reconfigurable intelligent surface (RIS)-assisted air-to-ground (A2G) communications with unmanned aerial vehicles (UAVs) in terrestrial and non-terrestrial networks under the impact of channel aging. We first model the G2A and A2G signal-to-noise ratios (SNRs) as non-central complex Gaussian quadratic random variables (RVs) and derive their exact probability density functions, offering a unique characterization for the A2G SNR as the product of two scaled non-central chi-square RVs. Moreover, we also find that, for a large number of RIS elements, the RIS-assisted A2G channel can be characterized as a single Rician fading channel. Our results reveal the presence of channel hardening in A2G communication under low UAV speeds, where we derive the maximum target spectral efficiency (SE) for a system to maintain a consistent required outage level. Meanwhile, high UAV speeds, exceeding 50 m/s, lead to a significant performance degradation, which cannot be mitigated by increasing the number of RIS elements.

This paper studies beam bandwidth partitioning problem in mobile millimeter-wave (mmWave) and multiple antennas networks. The main novelty is to flexibly optimize the beamforming bandwidth with the aim to minimize the energy consumption of the system while guaranteeing the data requirements of all mobile users. We formulate the problem as an integer nonlinear programming problem. To efficiently solve the problem, we design a deep reinforcement learning using the proximal policy optimization approach and train a deep neural network in an on-policy manner. Then, for comparison purposes, we develop low-complexity online iterative accurate solutions. We show that our approach achieves better performance compared to the iterative solutions and is able to achieve at least 4% less energy consumption and more than 12% energy efficiency gains.

This paper investigates a deep reinforcement learning (DRL)-based approach for managing channel access in wireless networks. Specifically, we consider a scenario in which an intelligent user device (iUD) shares a time-varying uplink wireless channel with several fixed transmission schedule user devices (fUDs) and an unknown-schedule malicious jammer. The iUD aims to harmoniously coexist with the fUDs, avoid the jammer, and adaptively learn an optimal channel access strategy in the face of dynamic channel conditions, to maximize the network's sum cross-layer achievable rate (SCLAR). Through extensive simulations, we demonstrate that when we appropriately define the state space, action space, and rewards within the DRL frame-work, the iUD can effectively coexist with other UDs and optimize the network's SCLAR. We show that the proposed algorithm outperforms the tabular Q-learning and a fully connected deep neural network approach.

Terahertz (THz) communication is a promising technology for future wireless communications, offering data rates of up to several terabits-per-second (Tbps). However, the range of THz band communications is often limited by high pathloss and molecular absorption. To overcome these challenges, this paper proposes intelligent reconfigurable surfaces (IRSs) to enhance THz communication systems. Specifically, we introduce an angle-based trigonometric channel model to evaluate the effectiveness of IRS-aided THz networks. Additionally, to maximize the sum rate, we formulate the source-IRS-destination matching problem, which is a mixed-integer nonlinear programming (MINLP) problem. To solve this non-deterministic polynomial-time hard (NP-hard) problem, the paper proposes a Gale-Shapley-based solution that obtains stable matches between sources and IRSs, as well as between destinations and IRSs in the first and second sub-problems, respectively.

In this paper, we present a novel coexistence uplink-downlink stochastic mobile edge computing (MEC) system that considers the dynamic characteristics of both small cell base stations (SBSs) and user equipments (UEs). To devise an efficient radio resource management strategy encompassing user association, channel allocation, and power allocation, we formulate an optimization problem that considers time, energy, and achievable rate in the utility function. The formulated problem is a Mixed Integer Nonlinear Program (MINLP) and has been proven to be NP-hard. To address this complexity, we propose a unified nature-inspired optimization framework, which can be deployed for subproblems in various settings and can be integrated with the Whale Optimization Algorithm (WOA), Improved Whale Optimization Algorithm (IWOA), and Particle Swarm Optimization (PSO). Through our rigorous mathematical and numerical analysis, the proposed algorithms show that they can converge to a near-optimal solution while keeping negligible optimality gaps. Our numerical results show the advantages and drawbacks of the proposed algorithms, highlighting their potential for effective resource management in MEC systems. The results also show the performance evaluation of stochastic characteristics on the performance of MEC.

In this paper, we propose federated learning (FL)-based jamming detection algorithms for a stochastic, distributed, tactical terrestrial and non-terrestrial (SDT-TNT) network. Specifically, we consider an SDT-TNT network with multiple clusters, in which multiple unknown jammers might be present. Moreover, we employ the spectral correlation function (SCF) on local servers to estimate the cyclostationary properties of the received waveforms. We then use the SCF to train local convolutional autoencoders (CAEs). In the inference phase, we jointly use the latent representation of the trained CAE and the kernel density estimation (KDE) to detect the existence of jammers. Our proposed methods show very promising results in jamming detection and outperform non-FL approaches. We further demonstrate that using the SCF feature provides higher accuracy than using In-phase/Quadrature-phase (I/Q) features.

In this paper, we propose a federated learning (FL)-based jammer detection and waveform classification algorithm for distributed tactical wireless networks (TWNs). More specifically, we consider a distributed TWN with multiple clusters and various types of waveforms used in the presence of a mobile jammer. We analyze the frequency domain of the waveforms received on local servers to extract the unique cyclic frequency from each waveform’s spectral correlation function (SCF). The method is used to detect the peak values in the frequency-cyclic frequency plane. The primary signal’s SCF exhibits peaks at the unique cyclic frequency and the zero cyclic frequency. These features are then used to train local convolutional neural networks (CNNs) to detect the jamming attacks and classify the waveforms. Moreover, a practical distributed TWN is considered in which each cluster head has a partial observation of the TWN with insufficient data samples, and the proposed algorithm exploits the distributed learning feature of FL, i.e., global learning aggregation to detect the existence of jammers and distinguish the types of waveforms received throughout the TWN. We implement a rigorous TWN simulation using MATLAB toolboxes and our proposed algorithm in TensorFlow Federated. The numerical results show that our proposed algorithm outperforms the standalone local SCF-CNN algorithm. We further demonstrate that the SCF feature yields more accuracy than the In-phase and Quadrature features.

In this paper, we aim to improve the connectivity, scalability, and energy efficiency of machine-type communication (MTC) networks with different types of MTC devices (MTCDs), namely Type-I and Type-II MTCDs, which have different communication purposes. To this end, we propose two transmission schemes called connectivity-oriented machine-type communication (CoM) and quality-oriented machine-type communication (QoM), which take into account the stochastic geometry-based deployment and the random active/inactive status of MTCDs. Specifically, in the proposed schemes, the active Type-I MTCDs operate using a novel Bernoulli random process-based simultaneous wireless information and power transfer (SWIPT) architecture. Next, utilizing multi-user power-domain non-orthogonal multiple access (PD-NOMA), each active Type-I MTCD can simultaneously communicate with another Type-I MTCD and a scalable number of Type-II MTCDs. In the performance analysis of the proposed schemes, we prove that the true distribution of the received power at a Type-II MTCD in the QoM scheme can be approximated by the Singh-Maddala distribution. Exploiting this unique statistical finding, we derive approximate closed-form expressions for the outage probability (OP) and sum-throughput of massive MTC (mMTC) networks. Through numerical results, we show that the proposed schemes provide a considerable sum-throughput gain over conventional mMTC networks.

In this paper, we propose and investigate an aerial reconfigurable intelligent surface (aerial-RIS)-aided wireless communication system. Specifically, considering practical composite fading channels, we characterize the air-to-ground (A2G) links by Namkagami-m small-scale fading and inverse-Gamma large-scale shadowing. To investigate the delay-limited performance of the proposed system, we derive a tight approximate closed-form expression for the end-to-end outage probability (OP). Next, considering a mobile environment, where performance analysis is intractable, we rely on machine learning-based performance prediction to evaluate the performance of the mobile aerial-RIS-aided system. Specifically, taking into account the three-dimensional (3D) spatial movement of the aerial-RIS, we build a deep neural network (DNN) to accurately predict the OP. We show that: (i) fading and shadowing conditions have strong impact on the OP, (ii) as the number of reflecting elements increases, aerial-RIS achieves higher energy efficiency (EE), and (iii) the aerial-RIS-aided system outperforms conventional relaying systems.

In this article, we study the physical-layer security (PHY-security) improvement method for a downlink nonorthogonal multiple access (NOMA) system in the presence of an active eavesdropper. To this end, we propose a full-duplex (FD)-enabled NOMA system and a promising scheme, called the minimal transmitter selection (MTS) scheme, to support secure transmission. Specifically, the cell-center and cell-edge users act simultaneously as both receivers and jammers to degrade the eavesdropper channel condition. Additionally, the proposed MTS scheme opportunistically selects the transmitter to minimize the maximum eavesdropper channel capacity. To estimate the secrecy performance of the proposed methods, we derive an approximated closed-form expression for secrecy outage probability (SOP) and build a deep neural network (DNN) model for SOP evaluation. Numerical results reveal that the proposed NOMA system and MTS scheme improve not only the SOP but also the secrecy sum throughput. Furthermore, the estimated SOP through the DNN model is shown to be tightly close to other approaches, i.e., the Monte-Carlo method and analytical expressions. The advantages and drawbacks of the proposed transmitter selection scheme are highlighted, along with insightful discussions.

In this paper, we study the statistical characterization and modeling of distributed multi-reconfigurable intelligent surface (RIS)-aided wireless systems. Specifically, we consider a practical system model where the RISs with different geometric sizes are distributively deployed, and wireless channels associated to different RISs are assumed to be independent but not identically distributed (i.n.i.d.). We propose two purpose-oriented multi-RIS-aided schemes, namely, the exhaustive RIS-aided (ERA) and opportunistic RIS-aided (ORA) schemes. A mathematical framework, which relies on the method of moments, is proposed to statistically characterize the end-to-end (e2e) channels of these schemes. It is shown that either a Gamma distribution or a Log-Normal distribution can be used to approximate the distribution of the magnitude of the e2e channel coefficients in both schemes. With these findings, we evaluate the performance of the two schemes in terms of outage probability (OP) and ergodic capacity (EC), where tight approximate closed-form expressions for the OP and EC are derived. Representative results show that the ERA scheme outperforms the ORA scheme in terms of OP and EC. In addition, under i.n.i.d. fading channels, the reflecting element settings and location settings of RISs have a significant impact on the system performance of both the ERA or ORA schemes.

This paper studies the performance of multi-hop cognitive wireless powered communication networks (WPCNs), where secondary nodes can harvest energy from multiple dedicated power beacons and access spectrum of primary receivers (PRs) to support data transmission. Specifically, we consider a practical scenario of cognitive WPCNs, where the channel state information (CSI) of interference links is assumed to be imperfect. To improve the network performance, we propose dual-hop scheduling scheme (DHS) for multi-hop network, where two best relays in two consecutive clusters are selected for data transmission. We then analyze the performance of the proposed scheme in terms of outage probability and outage floor. Numerical results show that DHS scheme outperforms random scheduling scheme, arising as an efficient scheme for multi-hop transmission. Furthermore, the effects of the number of hops, number of power beacons, and time switching ratio on multi-hop cognitive WPCNs are evaluated and discussed.

In this paper, we study the problem of jointly optimizing user pairing and beamforming design in multiuser multiple-input single-output (MU-MISO) non-orthogonal multiple access (NOMA) downlink systems with simultaneous wireless information and power transfer (SWIPT). Aiming at maximizing the achievable sum throughput subject to energy harvesting (EH) constraints, we propose a hybrid user pairing beamforming scheme (HBS), where two users with distinct channel conditions are optimally selected to perform user pairing. Moreover, we adopt a non-linear EH model for energy users to reveal their practical circuit characteristics. The sum throughput problem is formulated as a class of mixed-integer nonconvex optimization programming which is computationally prohibitive. To solve this challenging problem, we propose a low-complexity iterative algorithm, yet efficient, based on sequential convex approximation method to arrive at least the local optima. Numerical results are provided to demonstrate the performance improvement of the proposed HBS scheme over the multiuser beamforming one without user pairing, revealing to be an effective scheme for MU-MISO-NOMA downlink systems.

In this paper, we study the performance improvement methods for a cell-edge user of two-user non-orthogonal multiple access (NOMA) systems in downlink scenarios. To this end, we propose two cooperative relaying schemes, namely ON/OFF-full-duplex relaying (ON/OFF-FDR) and ON/OFF-half-duplex relaying (ON/OFF-HDR) schemes. More specifically, in order to improve the performance of the cell-edge user, we consider a cell-center user as a relay, where either FDR or HDR can be employed to assist the direct NOMA transmission from a base station (BS) to the cell-edge user. An ON/OFF mechanism is proposed to decide whether the cooperative relaying transmission is necessary or not. The ON/OFF relaying decision is made based on the quality of the direct and relaying links from the BS to the cell-edge user. The performance of the two proposed schemes is investigated in terms of outage probability and sum throughput. Numerical results reveal that the proposed schemes not only provide essential outage performance improvements for the cell-edge user, but also are able to improve the sum throughput of the two-user NOMA systems. The advantages and drawbacks of each proposed scheme are highlighted and insightful discussions are provided.

In this paper, we study performance and reliability improvement for a cell-edge user in downlink scenarios of two-user non-orthogonal multiple access (NOMA) networks. To this end, we propose a full-duplex (FD) cooperative scheme, in which a near user acts as a FD relay to forward source's signals to a far user, while the far user employs distributed switch-and-stay combining (DSSC) technique to process the incoming signals. We then investigate the performance of the far user in terms of outage probability (OP). In particular, we obtain a closed-form expression for the OP of the far user as well as its asymptotic OP. The developed analysis is corroborated through Monte-Carlo simulation. Numerical results reveal that the proposed scheme achieves better outage performance in comparison with conventional NOMA systems. It is also showed that the choice of the switching threshold used in DSSC technique and/or the target data rate of the system sensitively affects the outage performance of the proposed scheme.

In this paper, we investigate outage performance and diversity gain of transmit antenna selection (TAS) schemes in two-user multiple-input single-output non-orthogonal multiple access (MISO-NOMA) cooperative downlink transmissions. To this end, two TAS criteria, namely Criterion I and Criterion II, are proposed, which select an antenna that experiences the best fading condition of the channel from the source to the far user and to the near user, respectively. Additionally, considering the near user as a relay to help improve the reliability of the far user, hybrid simultaneous wireless information and power transfer (SWIPT) architecture is adopted to power the near user's relaying operation. Tight closed-form approximate expressions for the outage probability (OP) of both users are derived. Numerical results reveal that Criterion I and II achieve, respectively, the diversity order of K + 1 and 2 at the far user, and 1 and K at the near user, where K denotes the number of transmit antennas at the base station.

In this paper, we analyze the downlink outage performance of opportunistic scheduling in dual-hop cooperative networks consisting of one source, multiple radio-frequency energy harvesting relays, and multiple destinations. To this end, two low-complexity, suboptimal, yet efficient, relay-destination selection schemes are proposed, namely direct links plus opportunistic channel state information (CSI)-based selection (DOS) scheme and direct links plus partial CSI-based selection (DPS) scheme. Considering three relaying strategies, i.e., decode-and-forward (DF), variable-gain amplify-and-forward (VG-AF), and fixed-gain amplify-and-forward (FG-AF), the performance analysis in terms of outage probability (OP) is carried out for each selection scheme. For the DF and VG-AF strategies, exact analytical expressions and tight closed-form approximate expressions for the OP are derived. For the FG-AF strategy, an exact closed-form expression for the OP is provided. Additionally, we propose a gradient-based search method to find the optimal values of the power-splitting ratio that minimizes the attained OPs. The developed analysis is corroborated through Monte Carlo simulation. Comparisons with the optimal joint selection scheme are performed and it is shown that the proposed schemes significantly reduce the amount of channel estimations while achieving comparable outage performance. In addition, regardless of relaying strategy used, numerical results show that the DOS scheme achieves full diversity gain, i.e., M + K, and the DPS scheme achieves the diversity gain of M+1, where M and K are the numbers of destinations and relays, respectively.

In this study, the authors analyse the outage performance of multirelay decode‐and‐forward cooperative networks subject to two joint practical issues of wireless communications, namely, energy constraints and transceiver hardware impairment (HI). To deal with energy constraints at relay nodes, radio‐frequency (RF) energy harvesting (EH) technique is adopted. Considering the joint impacts of RF EH technique and HI, two relay selection schemes, namely, harvested energy‐based relay selection (HEbS) scheme and channel quality‐base relay selection (CQbS) scheme are proposed. Tight closed‐form approximate expressions for the outage probability of each scheme are derived and corroborated through Monte Carlo simulations. Some representative performance comparisons are carried out and show that the diversity order achieved by the HEbS scheme is 1 and is independent of the number of relays, K , whereas that achieved by the CQbS scheme is K , which is full diversity.