TD3 Algorithm-Based SWIPT With UAV-RIS Assistance for MIMO Communication

Unmanned aerial vehicles (UAVs) equipped with reconfigurable intelligent surfaces (RIS) provide several perquisites in inaccessible regions, yet the limited battery life of the UAV is a concern that needs to be addressed. This study suggests a new scheme that combines the nearest neighbor search (NNS) method for UAV trajectories and a simultaneous wireless information and power transfer (SWIPT) model that splits the passive reflecting elements of the RIS in the geometric space to forward information and harvest energy concurrently. Nonetheless, users' mobility and ever-changing channel conditions pose a challenge in achieving an optimal wireless system. To tackle these issues, this study employs a twin delayed deep deterministic policy gradient (TD3)-based algorithm to enhance the proposed SWIPT model while ensuring the QoS. Simulation results illustrate the efficiency of the proposed robust TD3-based SWIPT model with UAV-RIS assistance implemented in MIMO communication. The proposed model has demonstrated superior performance in comparison to existing solutions. The results show that the model achieves a 70% energy efficiency and improves the system's capacity by up to 56.3%.