Joint Power Control and Beamforming for Interference Mitigation in Multi-User Millimeter-Wave Systems

Interference poses a significant challenge in improving cell throughput in multi-cell multi-user networks. Coordinated beamforming and power control have shown promise in mitigating interference and maximizing cell throughput. However, existing techniques often suffer from computational complexity and the overhead of collecting channel state information (CSI) from interfering cells. In this paper, we propose a deep reinforcement learning based approach that addresses these limitations by eliminating the need for explicit CSI knowledge. Through extensive simulations in millimeter-wave networks with various cell configurations, we demonstrate the effectiveness of our interference management technique. The simulation results showcase the ability of our method to learn near-optimal power and beamforming strategies for multi-cell multi-user networks, all without the need for explicit CSI information.