The integration of autonomous aerial vehicles (AAVs) and reconfigurable intelligent surfaces (RISs) in sixth-generation (6G) networks offers significant potential for enhancing integrated sensing and communication (ISAC) systems. Motivated by the need for real-time adaptability, spectrum efficiency, and intelligent wireless infrastructure, this article investigates a AAV-RIS-enabled ISAC framework capable of meeting the dual demands of sensing and communication in complex environments. A key challenge in such systems lies in dynamically balancing resource allocation, AAV trajectory planning, and interference management. To address this, we propose a joint optimization framework in which a AAV simultaneously conducts sensing operations and serves multiple ground users in obstructed environments. The formulated problem involves a nonconvex tradeoff between maximizing sensing signal-to-noise ratio (SNR) and communication throughput. To solve it, we develop a multiagent deep reinforcement learning (MADRL) solution. Three collaborative agents independently optimize AAV trajectories using deep deterministic policy gradient (DDPG), RIS phase shifts using twin delayed DDPG (TD3), and beamforming matrices using proximal policy optimization (PPO). These agents learn optimal policies under dynamic channel conditions and mobility constraints. Simulation results show that the proposed approach improves sensing SNR by 22% and communication rates by 20% compared to baseline methods. With equal weight factors for sensing and communication, the system achieves 95% of the maximum theoretical sensing SNR while ensuring a minimum communication rate of 1 bps/Hz, validating its effectiveness in real-time ISAC environments.

In this paper, we present a transmission strategy optimization technique for multiple unmanned aerial vehicles (UAVs) base stations assisted by a terrestrial reconfigurable intelligent surface (RIS). While each UAV is designed to serve a particular group of users, whether a UAV is assisted by RIS depends on the distance between the UAV and the terrestrial RIS. As such, a joint optimization problem of the allocation of RIS clusters for each associated UAV and the computation of phase rotations corresponding to terrestrial RIS elements is formulated. Based on the generalized Benders decomposition (GBD), we develop an algorithm to solve the problem efficiently for this mixed integer non-linear programming problem (MINLP) in two cases: 1) treating interference as noise (TIN) and 2) applying successive interference cancellation (SIC) at user terminals. Through extensive simulations, we provide evidence of our algorithm's superiority over existing methods, including alternating optimization and multi-task learning. By checking the upper and lower bounds provided by the proposed algorithm, we show that our algorithm attains the global optimality for the considered simulations. Notably, our results also reaffirm the common belief that SIC decoding can provide performance enhancement over TIN decoding in the UAV-RIS-assisted systems.

This study explores the optimization of transmission strategies focusing on energy efficiency within a network comprising ground-based beyond-diagonal reconfigurable intelligent surfaces (BD-RIS), a.k.a RIS 2.0, and multiple unmanned aerial vehicles (UAVs). The motivation behind this work stems from the critical need to enhance energy efficiency in next-generation wireless networks, where the integration of UAVs and RIS technologies presents both opportunities and challenges. Specifically, while UAVs offer flexible deployment and improved coverage, their limited battery life and the complex interference environment in multi-user networks necessitate innovative solutions for sustainable operation. Each UAV is designed to serve its corresponding user group, with each group utilizing unique subcarriers to maintain orthogonality and employing a rate-splitting multiple access (RSMA) strategy within each group. The primary objectives of this work are to optimize: 1) the allocation of BD-RIS elements to groups, 2) the phase rotations of BD-RIS, 3) the common rate allocation in RSMA, 4) UAV trajectories, and 5) the design of precoders. To achieve these objectives, we formulate an optimization problem under the framework of mixed-integer nonlinear programming (MINLP), with a focus on maximizing energy efficiency. Our proposed solution combines generalized Benders decomposition (GBD), a manifold-based algorithm, and successive convex approximation (SCA). GBD decomposes the MINLP into primal and master sub-problems, which are iteratively solved. To efficiently address variable coupling in the primal problem, we adopt a block coordinate descent (BCD) method and employ the Riemannian conjugate gradient (RCG) technique for phase rotation. SCA addresses the remaining challenges in the primal problem, while a two-stage approach simplifies the optimization process. Simulations confirm the significant energy efficiency improvements achieved by the proposed method.

Reconfigurable intelligent surfaces (RIS) is a new technology that aims to improve the spectral and energy efficiency for next-generation wireless technology (6G). It has recently emerged as a potential alternative for alleviating limitations caused by complex propagation environment, especially in urban areas and its ability to expand network coverage. The advantage of RIS is especially evident when line-of-sight transmission is not possible in aerial-terrestrial communication where unmanned aerial vehicles (UAVs) are used as access points (APs). In most literature, the RIS is shared among different access points (APs) and the phase shifters are designed jointly to optimize certain performance metric for all APs and receivers. This work takes a different approach by optimally determining which RIS elements should be assigned to each UAV-user pair and designing the phase shifters at the RIS to maximize the sum rate accordingly. The problem is formulated as a mixed-integer nonlinear programming problem and solved using the Generalized Benders Decomposition (GBD) method. Simulation results show that substantial performance gain is obtained using the proposed strategy compared to non-grouping scheme.

Multiuser-multiple input multiple outputs (MUMIMO) can provide a substantial performance for cellular networks by applying a beamforming technique to direct signals in specific required directions. Therefore, this paper presents a basic idea of steering the incoming signals into the signal of interest (SOI) and signal not of interest (SNOI) directions. Moreover, different adaptive beamforming algorithms such as least mean square (LMS), recursive least square (RLS) and constant module algorithm (CMA) are analyzed and evaluated regarding the interference cancelation and steering of incoming data signal. In this regard, to calculate the received output signal, the optimal weight elements of a uniform linear array (ULA) antenna are computed and updated based on the incoming array sensor signals. Furthermore, the paper shows the effect of the beamforming on a bit error rate (BER) when applied a different number of antenna elements at base station (BS). The simulation results confirm that the algorithms likely have the same performance in the interference cancelation. However, the RLS algorithm shows a little enhancement compared with SMI and CMA algorithms. Besides, the results indicate that the beamforming with the different number of antenna elements has a significant impact on the BER performance. Hence, this work can be extended to a higher frequency of mmWave communication with a massive number of antenna elements, which can meet requirement of fifth generation (5G) network systems.

Multiuser-multiple input multiple output (MU- MIMO) systems can provide a substantial performance for cellular networks by applying direction of arrival (DOA) technique, in which the direction of sidelobe signals can be estimated. Therefore, this paper presents a different adaptive DOA algorithms where the performance of communication networks can be enhanced, in the meantime the minimizing interferences. Meanwhile, The DOA algorithm such as multi signal classification (MUSIC), minimum norm (Min-Norm), minimum variance distortionless response (MVDR) and Bartlett algorithms are studied and compared extensively regarding the estimation angles and peak power spectrum. In this regard, uniform linear array (ULA) antenna is employed to construct the weight vector for incoming sensor signals. Furthermore, this paper compares the DOA algorithms concerning the number of antenna elements as well as the incoming signal angles. The simulation results approve that the algorithms are likely have the same performance in accuracy of scanning angles. However, it is realized that all algorithms have different peak power values in order to beamwidth. Moreover, the results indicate that the DOA algorithms with the different number of antenna elements have a significant impact on a directivity and gain. Hereafter, this work can be extended to higher frequency of mmWave communication with massive number of antenna, which can meet the requirement of fifth generation (5G) network.