Abstract
In the multi-TRP (Transmission Reception Point) networks, the user equipment (UE) suffer the interference from interfering TRPs. However, Coordinated Multi-Point Joint Transmission (CoMP-JT) technique is proposed to solve the problem, which could transfer interfering signals into useful signals. As CoMP-JT employed, data for a UE is jointly transmitted from multiple TRPs in the same cooperating cluster, thus improve the UEs’ data rate. In this chapter, we investigate the joint downlink power allocation and user association for the deployment of multi-TRP networks with CoMP-JT for maximizing the system spectral efficiency. Since the problem of joint user association and power allocation is NP-hard and thus hard to solve, we develop a deep reinforcement learning approach due to its capability to provide approximate solutions while dealing with a large-scale problem. The parallel solution of objective function is considered in the proposed deep Q-learning (DQL). Furthermore, we also establish a two-stage optimal approach, where the objective function is decomposed into two sub-problems, i.e., user association and power allocation. Both sub-problems are solved by DQL. Simulation results are provided to evaluate whether the proposed method can achieve near-optimal solutions compared with the other benchmark algorithms and also demonstrate that DQL can provide good performance in each sub-problem.