Multi RIS-Integrated MIMO Wireless Communication: End to End (E2E) Learning Performance across Diverse Network Scenarios

Reconfigurable Intelligent Surfaces (RIS) have emerged as a transformative technology, revolutionizing wireless communication systems by enabling dynamic manipulation of radio waves. Integrating this capability, the End-to-End (E2E) learning approach amalgamates RIS into the wireless communication paradigm, offering adaptive and efficient communication in diverse scenarios. In this study, we present an extensive evaluation and comparative analysis of the E2E learning approach’s performance in various RIS-empowered scenarios. Expanding upon the previously introduced communication model, we incorporate a range of new scenarios involving different quantities of RISs and various User Equipment (UE) configurations. These scenarios encompass the deployment of multiple RIS units catering to individual UEs, as well as the presence of multiple UEs. Through meticulous examination of simulation results, we gain profound insights into the adaptability, robustness, and efficiency of the E2E learning approach under distinct RIS-aided scenarios.