Interference Mitigation in STAR-RIS-Aided Multi-User Networks with Statistical CSI

In this paper, we investigate real-time interference mitigation in multiuser wireless networks assisted by simultaneously transmitting and reflecting reconfigurable intelligent surfaces (STAR-RISs). Unlike conventional methods that rely on instantaneous channel state information (CSI), we consider a practical scenario where only statistical CSI is available, and the STAR-RIS phase shifts are impaired by random phase errors modeled via the Von Mises distribution. To tackle the resulting nonconvex optimization problem induced by unit-modulus constraints and stochastic interference, we derive a closed-form approximation of the effective channel matrix using statistical expectations. We then reformulate the interference minimization problem as an unconstrained optimization over a Riemannian manifold and propose a conjugate gradient algorithm tailored to the complex circle manifold. The proposed solution enables efficient real-time computation of optimal phase shifts while accounting for hardware imperfections and limited CSI. Simulation results confirm that our method significantly suppresses inter-user interference and achieves superior SINR performance and convergence speed compared to conventional baselines.

Adaptive Soft Actor-Critic Framework for RIS-Assisted and UAV-Aided Communication

In this work, we explore UAV-assisted reconfigurable intelligent surface (RIS) technology to enhance downlink communications in wireless networks. By integrating RIS on both UAVs and ground infrastructure, we aim to boost network coverage, fairness, and resilience against challenges such as UAV jitter. To maximize the minimum achievable user rate, we formulate a joint optimization problem involving beamforming, phase shifts, and UAV trajectory. To address this problem, we propose an adaptive soft actor-critic (ASAC) framework. In this approach, agents are built using adaptive sparse transformers with attentive feature refinement (ASTAFER), enabling dynamic feature processing that adapts to real-time network conditions. The ASAC model learns optimal solutions to the coupled subproblems in real time, delivering an end-to-end solution without relying on iterative or relaxation-based methods. Simulation results demonstrate that our ASAC-based approach achieves better performance compared to the conventional SAC. This makes it a robust, adaptable solution for real-time, fair, and efficient downlink communication in UAV-RIS networks.