Speaker: Luis Castedo, Department of Computer Engineering, University of A Coruña, Spain.
Abstract: The convergence of multiple-input multiple-output (MIMO) systems and intelligent reflecting surfaces (IRSs) is anticipated to play a crucial role in enabling beyond 5G (B5G) and 6G technologies. This talk explores the joint optimization of the IRS phase-shift matrix and MIMO precoders of an IRS-assisted multi-stream (MS) multi-user MIMO (MU-MIMO) system, with the goal of maximizing system sum-rate for every channel realization. Conventional optimization approaches prove inadequate for this intricate task. Instead, we turn to deep reinforcement learning (DRL), which, unlike supervised learning, trains by interacting directly with the communication system itself. We will elucidate how to adapt the DRL paradigm to effectively address this optimization problem and demonstrate its ability to outperform state-of-the-art heuristic methods in scenarios with high multi-user interference.