This project focuses on developing an ultra-massive MIMO IRS-assisted two-way communication system in the mmWave spectrum. It integrates Multi-Agent Reinforcement Learning (MARL) for real-time, collaborative channel estimation, improving system performance. Additionally, a MARL-based beamforming module will optimize beam alignment, minimizing misalignment errors and enhancing Quality of Service (QoS) using channel state information.
Accessibility Controls