Service Fairness Enhancement for BDRIS Assisted Fluid Antenna Systems

The rapid surge in network sizes, driven by the proliferation of wireless applications, has placed unprecedented demands on wireless systems leading to high interference, spectrum bottlenecks, and increased power utilization. Reconfigurable intelligent surfaces (RISs) have emerged as a promising candidate for these challenges owing to their passive beamforming action, but have limited gains due to independently functioning elements. Consequently, beyond-diagonal RISs (BDRISs) are proposed as the key-enabler of next-generation systems. In this paper, we propose a fairness-aware design for BDRIS-assisted fluid antenna systems. We jointly optimize the antenna position vector, beamforming vectors, and BDRIS configuration matrix to address service fairness while meeting the power budget in a multi-user downlink system. To tackle this non-convex problem, we utilize proximal policy optimization and obtain a fast-converging solution. Our results demonstrate a 2.8 bps/Hz mean rate improvement as compared to conventional RIS systems.