Aerial Reconfigurable Intelligent Surface-Assisted Secrecy Communication Based on SWIPT

In this paper, we propose a high-energy-efficiency secrecy communication method using an aerial reconfigurable intelligent surface (ARIS) to assist secrecy communication in environments with communication barriers. Specifically, we decompose the reflective array of the meta-surfaces in the geometric domain for simultaneous wireless information and power transfer (SWIPT). The phase shift matrix of the metasurfaces is then optimized to maintain the security of the communication. To further improve system performance, we introduce a proximal policy optimization (PPO) approach in deep reinforcement learning (DRL) that allocates ARIS resources continuously in both the time and space domains, ensuring secure communication while minimizing overall energy consumption through efficient energy harvesting. Simulation results validate the effectiveness of the proposed method, demonstrating that it not only ensures secrecy communication quality but also achieves significant energy efficiency in the process.