Deep Reinforcement Learning for 5G Networks: Joint Beamforming, Power Control, and Interference Coordination
The fifth generation of wireless communications (5G) promises massive increases in traffic volume and data rates, as well as improved reliability in voice calls. Jointly optimizing beamforming, power control, and interference coordination in a 5G wireless network to enhance the communication performance to end users poses a significant challenge. In this paper, we formulate the joint design of beamforming, power control, and interference coordination to maximize the signal to interference plus noise ratio (SINR) and solve the non-convex problem using deep reinforcement learning. By using the greedy nature of deep Q-learning to estimate future benefits of actions, we propose an algorithm for voice bearers in sub-6 GHz bands and data bearers in millimeter wave (mmWave) frequency bands. The algorithm exploits reported SINR from connected users, the transmit powers of the base stations, and the coordinates of the connected users to improve the performance measured by coverage and sumrate capacity. The proposed algorithm does not require the channel state information and removes the need for channel estimation. Simulation results show that our algorithm outperforms the link adaptation industry standards for sub-6 GHz voice bearers and approaches the optimal limits for mmWave data bearers for small antenna sizes in realistic cellular environments.
READ FULL TEXT