A Learning-Based Approach to Joint UAV Trajectory and Beamforming Optimization for UAV-RIS Relaying Network

The performance benefits of reconfigurable intelligent surfaces (RISs) in enhancing communication capability and coverage are constrained by the static placement of the RIS. However, aerial-RIS, where the RIS is deployed on an unmanned aerial vehicle (UAV), has demonstrated superior potential to improve system performance due to the mobility and adaptability of the UAV. In this paper, we investigate a UAV-mounted RIS (UAV-RIS) relaying system and formulate an optimization problem aimed at maximizing the system sum rate. To address this, we propose a novel algorithm named DL-UTBO, that integrates deep reinforcement learning with a deep neural network (DNN) to jointly optimize the UAV trajectory, base station (BS) active beamforming, and RIS passive beamforming. The simulation results underscore the effectiveness and superiority of the proposed algorithm compared to existing baseline methods.