DRL-Based Dynamic Resource Allocation for Multi-Beam Satellite Systems

Multi-beam satellite communication systems have received considerable attentions in recent years. By generating multiple beams at the transmitting satellites, the coverage areas can be enlarged and the transmission capacity can be improved. In this paper, we study dynamic radio resource allocation problem for multi-beam satellite systems. Considering the dynamic arriving of data packets and signal distortion due to nonlinear amplifying, we formulate the joint beam scheduling, subchannel and power allocation problem of the satellite system as a long-term average utility function maximization problem. To solve the formulated problem, we first propose a cell grouping and beam scheduling scheme. Then, we model the dynamic resource allocation problem of individual cells as a Markov decision process and propose a double deep Q-network (DDQN)-based subchannel and power allocation algorithm. Simulation results demonstrate the effectiveness and superiority of the proposed algorithm.