In this letter, we study max–min rate optimization in beyond-diagonal reconfigurable intelligent surface (BD-RIS)-aided uplink systems under channel aging with time-varying channels. Unlike existing studies, the channel update interval is jointly optimized with system resources to balance the trade-off between training overhead and channel aging effects using a Bayesian optimization framework. Globally optimal solutions are obtained for power allocation and beamforming subproblems, while near-optimal performance is achieved for BD-RIS, with the beamforming derived in closed form. Results reveal that both very short and long update intervals reduce performance and that the proposed approach significantly outperforms stochastic search over 107 configurations, and that the proposed BD-RIS solution is close to the relaxed global optimum for a small number of elements.

Cell-free multiple-input multiple-output (CF-MIMO) systems assisted by reconfigurable intelligent surfaces (RISs) and rate-splitting multiple access (RSMA) are promising for interference management in distributed sixth-generation (6G) networks. In practical CF-MIMO deployments, independent access-point (AP) oscillators and unequal propagation delays introduce AP-dependent phase variations that reduce coherent combining and degrade precoding performance. This paper develops a unified RIS-RSMA optimization framework for asynchronous downlink CF-MIMO. The RSMA common precoder, private precoders, and passive RIS phase shifts are jointly optimized within a sum-rate maximization problem subject to total transmit-power, per-user quality-of-service, and RIS unit-modulus constraints. The resulting non-convex problem is addressed through a weighted minimum mean-square error (WMMSE)-based alternating optimization framework with Gauss–Seidel coordinate optimization for the RIS phase subproblem. An analytical common-stream power expression is derived to explain the limited benefit of decoupled common-precoder designs. The analysis shows that independent oscillator phase drifts at the APs attenuate coherent cross-AP combining terms and can restrict the common rate through the weakest-user decoding constraint. In the default overloaded CF-MIMO configuration, the tested heuristic common-precoder schemes provide less than 0.2% gain over the corresponding optimized space-division multiple access (SDMA) baseline when combined with WMMSE-optimized private precoders. In contrast, the proposed joint RSMA design achieves approximately 12–15% sum-rate gain at moderate-to-high transmit powers and retains a similar high-power advantage in a larger-scale validation. Targeted comparisons using WMMSE and regularized zero-forcing (RZF) private precoders, together with the common-power allocation results, support the need for joint common/private precoder design. Additional results show that the gain persists across the tested RIS sizes, oscillator phase-noise variances, Rician factors, loading ratios, channel realizations, and multi-RIS deployments. Distributed multi-RIS deployment also improves spatial coverage uniformity under a fixed total element budget.

This paper investigates the optimal placement of a millimeter-wave (mmWave) base station (BS) within a realistic U-shaped environment with non-convex topology. The problem is challenging and NP-hard due to the non-convex topology and the non-convex objective functions which are the sum-rate maximization and max-min fairness, the latter being additionally non-smooth. To address this challenge, the BS placement is formulated as a Markov Decision Process (MDP). Then, we propose two deep reinforcement learning (DRL) techniques: First, the deployment area is discretized into a grid and optimized using a Deep Q-Network (DQN). Second, the U-shaped region is partitioned into continuous subspaces, where a Deep Deterministic Policy Gradient (DDPG) agent is dedicated to each subspace then the best BS placement is selected among partitions. Results demonstrate that optimal placement achieves full coverage and yields a Jain index of 0.99. Furthermore, the proposed partitioned multi-space DDPG achieves better solution than DQN with lower complexity.

Active Reconfigurable Intelligent Surface (RIS) is a promising technology for future wireless networks. With 6G research going ahead, the mutual deployment of RIS and relay has received increasing attention recently. In this work, we investigate the performance of a multi-active RIS-aided multi-user uplink system assisted by intermediate amplify and forward (AF) relays. We first focus on the double-active scenario with a single intermediate AF relay and jointly optimize the users' transmit powers, AF relay gain, the two active RISs reflecting elements, and base station (BS) beamforming to maximize the minimum rate. Due to the non-convexity, non-smoothness of the objective function, and an extremely huge number of optimization variables, we propose a hybrid optimization framework where users' power and active RISs elements are optimized using a deep reinforcement learning (DRL) approach, while the optimal relay gain and beamforming at the BS are obtained in closed-form. The proposed soft actor-critic (SAC) DRL agent is bench-marked against its counterparts agents from the literature, specifically, deep deterministic policy gradient (DDPG), and twin delayed DDPG (TD3). We then investigate the system performance under multiple active RISs and AF relays. Numerical results show the superiority of SAC compared to DDPG and TD3 with even lower computational cost. Furthermore, the allocation of users' power and first RIS are the most effective key performers for the proposed chain of active RISs and AF relays meanwhile the relay gain, BS beamforming, and following RISs have much smaller impact on the proposed system performance.

Sixth-generation (6G) wireless systems require advanced multiple access techniques capable of delivering high data rates, low latency, and robust performance under practical impairments. Rate-Splitting Multiple Access (RSMA) has emerged as a strong candidate due to its efficient interference management and robustness to imperfect Channel State Information (CSI). However, its performance is often limited by the weakest user, and existing cooperative approaches mainly rely on half-duplex relaying and ideal assumptions. In this paper, a downlink full-duplex multi-user 2-layer cooperative RSMA (C-RSMA) framework is proposed under asynchronous reception, imperfect CSI, and imperfect Successive Interference Cancellation (SIC). The 2-layer structure enhances interference mitigation and fairness, while full-duplex relaying improves spectral efficiency. An alternative optimization technique based on Weighted Minimum Mean Square Error (WMMSE) is used to jointly optimize precoding, rate allocation, and relay power to maximize the minimum user rate under latency constraints. Numerical results show that the proposed scheme enhances fairness and robustness over other schemes.

This paper investigates the wireless resource allocation (RA) problem of maximizing the sum-rate for a multi-user active RIS-assisted uplink model. To reach the highest possible the sum-rate, we optimize three system resources which are the beamforming at the base station (BS), where we consider a multi-antenna BS, the users' transmit power, and the active RIS elements configurations including the passive phase shifts, and active gains. These resources are optimized in presence of imperfect channel state information (CSI) using a practical three-stage estimation technique. To solve this challenging non-convex RA problem, we propose a decomposition strategy where the original problem is decomposed into sub-ones. Numerous approaches are used as the Lagrangian method, fractional programming, the projected gradient descent, and first-order multi-variate Taylor approximation. The numerical analysis shows that active RISs meanwhile being superior, are revealed to be more sensitive to CSI imperfections especially when CSI error variance is high.

In this paper, the problem of maximizing the sum-rate is addressed for a multi-user uplink scenario that is assisted by an active reconfigurable intelligent surface (RIS). The maximization is achieved by optimizing the beamforming at the base station, the users' transmit power, active RIS elements phase shifts, and active gains in presence of imperfect channel state information (CSI). The non-convex maximization problem is decomposed into sub-problems and solved via iterative approaches including the Lagrangian method, the projected gradient descent, multi-variate Taylor expansion and fractional programming. Numerical results show that the active RIS is more sensitive to CSI imperfections than passive one at high error variances.

In this paper, we address the problem of optimizing power allocation of a rate-splitting multiple access (RSMA) system. The optimization task is formulated to maximize the RSMA sum-rate of a multi-user downlink scenario under a total power budget constraint. To deal with this non-convex objective, we compare the performance of three different approaches: the gradient descent (GD) algorithm, deep reinforcement learning (DRL) with deep deterministic policy gradient (DDPG), and DRL with soft actor-critic (SAC). Gradient descent is used as a traditional optimization technique, while DRL with DDPG/SAC represents state-of-the-art DRL techniques capable of handling complex and non-convex problems. Numerical results show that DDPG achieves the highest sum-rate performance, followed by GD and then SAC.

Active Reconfigurable Intelligent Surfaces (RIS) are a promising technology for 6G wireless networks. This paper investigates a novel hybrid deep reinforcement learning (DRL) framework for resource allocation in a multi-user uplink system assisted by multiple active RISs. The objective is to maximize the minimum user rate by jointly optimizing user transmit powers, active RIS configurations, and base station (BS) beamforming. We derive a closed-form solution for optimal beamforming and employ DRL algorithms: Soft actor-critic (SAC), deep deterministic policy gradient (DDPG), and twin delayed DDPG (TD3) to solve the high-dimensional, non-convex power and RIS optimization problem. Simulation results demonstrate that SAC achieves superior performance with high learning rate leading to faster convergence and lower computational cost compared to DDPG and TD3. Furthermore, the closed-form of optimally beamforming enhances the minimum rate effectively.

The sixth generation (6G) of wireless networks demands high-performance communication solutions to achieve massive connectivity and ultra-low latency services. To achieve high throughput, rate-splitting multiple access (RSMA) offers interference management compared to conventional schemes. On the other hand, unmanned aerial vehicle (UAV) in heterogeneous network (HetNet) provides high coverage, improves throughput, and achieves massive connectivity. In this paper, we integrate UAV assisted HetNet with RSMA, and user clustering to maximize system sum rate in the presence of eavesdroppers while guaranteeing power constraints. The optimization problem jointly optimizes the common and private rates along with beamforming. The ptoblem is inherently non convex and solved using successive convex approximation (SCA) and first order Taylor expansion. The preformance of the RSMA-based scheme is measured in terms of the sum rate and is compared with NOMA based system. Numerical results show that RSMA-based system outperform conventional NOMA systems across various configurations. In particular, the results show higher achievable sum rates with increased number of antenna, and robustness to varying UAV transmit power and cluster separation distances.

Rate splitting multiple access (RSMA) is one of the promising multiple access techniques in 6G. One of the limitations of RSMA is that the rate is limited by the worst user’s rate. Cooperative rate splitting multiple access (C-RSMA) has become an essential technique to overcome this limitation. In this work, a downlink full-duplex C-RSMA with imperfect channel state information (ICSI) and imperfect successive interference cancellation (ISIC) is considered. Full-duplex (FD) relaying is used to fully utilize the channel in order to enhance the system performance. The objective is to maximize the minimum rate of the users, while ensuring that the transmission latency for each user’s rate does not exceed a specified threshold. Success convex approximation (SCA) is used to optimize the precoding matrix, common rate allocation, and the relay power. Numerical results show that the proposed system improves the Max-Min rate compared to the other systems.

Recently, unmanned aerial vehicles (UAVs) and reconfigurable intelligent surface (RIS) have been used to enhance the performance of wireless communications. Signal transmission supported by the UAV can have a direct link with mobile users due to its flexible deployment and controllable trajectory. Non-orthogonal multiple access (NOMA) achieves high spectral efficiency by utilizing power domain multiplexing in a multiuser NOMA system. In this paper, a heterogeneous network (HetNet) is considered with a UAV acting as a small base station (BS) that serves terrestrial users in a microcell. The BS in the macro cell serves multi-users using NOMA technology with the aid of RIS. A joint optimization problem is formulated to optimize the allocated powers of all users and the phase shifts of the RIS to maximize the sum rate. The allocated power is optimized and closed form expressions are obtained using lagrangian dual transform, while the RIS phase shifts are optimized using the full search method. Numerical results show that the joint optimization of RIS phase shifts and power has a higher sum rate than optimizing the allocated power only or optimizing phase shifts only.

The Fifth Generation (5G) New Radio (NR) wireless system is the most promising next-generation solution to meet the needs of the increasing demands of mobile market. Orthogonal Frequency Division Multiplexing (OFDM) is the fundamental transmission technique due to the great improvement in spectral efficiency and the flexibility in fast varying environments. The use of multiple-input-multiple-output (MIMO) OFDM and Massive MIMO (Ma-MIMO) OFDM promises the increase in quality, throughput, and capacity of the communication system. The use of Deep Learning techniques in wireless communication networks has shown significant achievements to overcome the rising challenges as compared to the conventional solutions. In this paper, first we give a brief background on OFDM, MIMO-OFDM and Ma-MIMO. Second, we provide a survey on the use of deep learning techniques in OFDM receivers. We categorize the state-of-the-art into Block based and End-to-End deep learning models. Third, we provide a survey on the use of deep learning techniques in MIMO-OFDM and Ma-MIMO-OFDM. We categorize the state-of-the-art as model driven and data driven deep learning models. Finally, we present future research directions.

In the pursuit of an efficient 6G network that achieves an enhanced capacity with minimal power consumption, reconfigurable intelligent surfaces (RIS) and cell-free Multiple-Input-Multiple-Output (CF-MIMO) emerge as two key technologies. In this paper, a joint precoding algorithm is designed for a multi-RIS aided CF-MIMO network, where the objective is to maximize the weighted sum rate (WSR) in the presence of imperfect channel state information (CSI) at the users. This optimization problem is divided into subproblems to alternately design the access point (AP) precoders and the RISs' phase shift matrices. A feasible solution is then obtained through a combination of Lagrangian Dual Reformulation (LDR), Fractional Programming (FP), and multidimensional complex quadratic transform (MCQT).

The receiver’s architecture is always a critical issue in multiple-input multiple-output (MIMO) systems, as it has a great impact on the system’s performance and complexity. Accordingly, in this paper, a spectral decomposition-based minimum mean squared error (MMSE) receiver is proposed in an uplink multi-user MIMO (MU-MIMO) system with a full-duplex decode and forward relay. Imperfect channel state information is assumed at the relay and base station where, least-squares channel estimation technique is applied due to its simplicity and low complexity. An equivalent relay is applied for self-interference cancellation. While, multi-user interference is mitigated by using the block diagonalization concept. A closed-form expression for the mean squared error (MSE) of the estimated data symbol is derived in the presence of channel estimation error. Moreover, an optimization problem is solved to derive an expression for the proposed MMSE receiver’s matrix. Finally, the performance of the proposed receiver is compared to that of the conventional receiver and Zero-forcing (ZF) receiver. Results show the superiority of the proposed receiver over both receivers.

Rate splitting multiple access (RSMA) has emerged as a powerful multiple access technique for wireless communications. In this paper, RSMA is integrated with cooperative relaying and reconfigurable intelligent surfaces (RIS) to enhance the spectral efficiency of multi-user communication. We propose a selective approach that chose some users to decode the common stream using successive interference cancellation (SIC) and treat the private streams as noise. While the other users do not decode the common stream, they treat it as noise. The selection is based on the condition of the users' channel. The user with the worst channel condition is selected to treat the common stream as noise. The performance of the proposed system is measured in terms of Max-Min rate of the users. Alternative optimization (AO) is used to jointly optimize the time slot allocation, precoding matrix, common rate allocation, and phase shifts of RIS in the direct and cooperative phase iteratively. Numerical results show that the proposed selective cooperative rate splitting assisted by RIS (CRS-RIS) improves significantly the Max-Min rate when compared with other existing systems.

Reconfigurable Intelligent Surface (RIS) enabled wireless communications with Non Orthogonal Multiple Access (NOMA) is a promising technology for the next generation mobile communications. Millimeter wave (mmWave) communication has high frequency ranges from 30 to 300 GHz and supports giga-bit per second data rates. The path loss attenuation is very high at high frequencies compared to low frequencies. In this paper, we consider a downlink mmWave MIMO-NOMA cellular system aided by RIS where the base station is mounted with multiple antennas and multiple single antenna users. RIS phase shifts and users power are optimized to maximize energy efficiency such that the rate of each user exceeds a certain threshold. The optimization problem is a non-convex problem, which can be solved using Dinkelbach's algorithm with fractional programming. The maximization problem is converted to Quadratic Constraint Quadratic programming (QCQP), and the Lagrange augmented method is applied to get the optimum RIS phases. Numerical results show that all users satisfy the rate constraint under optimal power allocation.

Large Intelligent Surfaces are essential components in upcoming 6G networks due to their impressive capabilities and energy efficiency. To enhance network coverage, we introduce relays that can overcome distance and obstacle limitations. The focus is on integrating Large Intelligent Surfaces with amplify-and-forward relay techniques to facilitate communication between a base station and a full-duplex user, even in the presence of co-channel interference from device-to-device pairs nearby. To optimize data transmission, closed-form expression is derived that calculates the optimal power required to achieve maximum data rates while considering challenges like self-interference and co-channel interference from device-to-device communications. Also, quadratic transform for the non-convex problem is derived analytically to be able to solve the problem using CVX as well. Additionally, a neural network approach is proposed that accurately predicts the minimum power needed to achieve the maximum data rates. This neural network model proves to be a promising alternative with much less complexity.

In the domain of wireless communication, Full-Duplex Multiple-Input Multiple-Output (FD-MIMO) systems have emerged as a promising technology for achieving higher spectral efficiency and increased data rates. Conventional FD-MIMO systems employ complex receivers for channel estimation, self-interference cancellation, and precoding. However, these conventional receivers often struggle to adapt to dynamic channel conditions and impose computational challenges. This paper utilizes Convolutional Neural Networks (CNNs) as a replacement for traditional receivers in FD-MIMO systems. The proposed CNN-based receiver demonstrates adaptability and robustness, enabling it to handle the challenges of channel estimation, self-interference cancellation, and precoding in real-time. Our results show the performance of the CNN-based receiver, which significantly outperforms a conventional receiver that uses Kalman Filter for channel estimation and Zero Forcing for pre-coding in terms of bit error rate (BER) and overall system efficiency. The proposed receiver enhanced the average BER by 43 % as compared to the conventional receiver. Results have shown that the proposed receiver is robust against pilot contamination due to synchronization errors. The use of convolutional neural networks is a promising approach to replace the conventional receiver as it has the ability to learn and mitigate the effect of changes in the environment.

To accommodate the growth of data traffic of 6G and beyond networks, achieving a significant improvement in spectrum efficiency is inevitable. Full-duplex systems are very promising since they have the potential of increasing the spectral efficiency compared to half-duplex systems. The main challenge facing the deployment of full-duplex systems is self-interference. Conventional receivers include series of processing blocks that recover the desired signal after removing the effect of self-interference. In the presence of noise and fading channels, the received signal becomes distorted and the recovery is thus challenging. Deep learning algorithms have shown great success in efficient parameter estimation, as well as adaptive decision making. In this study, a deep learning-based full-duplex receiver, namely FDDR, is proposed to replace the receiver’s entire information recovery process of the full-duplex system rather than optimizing each processing module of the receiver separately. Simulation results show that FDDR approaches the same Bit Error Rate performance as a conventional receiver that uses a Kalman Filter as a channel estimator with 80% less complexity. The Bit Error Rate performance of FDDR is also tested in the case of Doppler frequency mismatch between the training and testing phases, multiple self-interference reflections and cyclic prefix free scenarios and shows superior performance.