Dynamic Resource Allocation With Deep Reinforcement Learning in Multibeam Satellite Communication

In this letter, the radio resource optimization in multibeam geostationary earth orbit (GEO) satellite communication (Satcom) is studied. We propose a deep reinforcement learning (DRL) algorithm based on the state-of-the-art twin delayed deep deterministic policy gradient (TD3) to jointly allocate the subchannel and power. Then we integrate independent training, prioritized experience replay, scaling factor, and noise rebound to address the bound action problem of TD3. Simulation results show that the proposed DRL-based algorithm outperforms the baseline schemes in terms of the sum log spectral efficiency.