Deep Reinforcement Learning for Optimizing RIS-Assisted HD-FD Wireless Systems

This letter investigates the reconfigurable intelligent surface (RIS)-assisted multiple-input single-output (MISO) wireless system, where both half-duplex (HD) and full-duplex (FD) operating modes are considered together, for first time in literature. The goal is to maximize rate by optimizing RIS phase shifts. A novel deep reinforcement learning (DRL) algorithm proposed solve formulated non-convex optimization problem. complexity analysis Monte Carlo simulations illustrate that DRL significantly improves compared non-optimized scenario HD FD using a single parameter setting. Besides, it reduces computational of downlink MISO system achievable with reduced number steps per episode conventional algorithm.