Joint 3D Deployment and Power Allocation for UAV-BS: A Deep Reinforcement Learning Approach

• DOI: 10.1109/lwc.2021.3100388
• Publication Date: 2021-07-27T20:31:50Z
• Authors: Meng Zhang; Shu Fu; Qilin Fan
• Keywords: Q-learning

Due to its high mobility and low cost, unmanned aerial vehicle mounted base station (UAV-BS) can be deployed in a fast cost-efficient manner for providing wireless services areas where traditional terrestrial infrastructures cannot laid technical economic reasons. In this letter, we investigate the problem of joint three-dimensional (3D) deployment power allocation maximizing system throughput UAV-BS system. To solve non-convex problem, propose deep deterministic policy gradient (DDPG) based algorithm. The proposed algorithm allows explore continuous state action spaces learn optimal 3D hovering location allocation. Simulation results show that outperforms Q-learning-based method genetic