查看文章

This research studies an uplink multiple-input multiple-output (MIMO) system with rate-splitting multiple access (RSMA) techniques. In this regard, we formulate a problem to maximize the achievable system sum rate by optimizing the precoding matrices at the users and the decoding order of all sub-messages received at the base station (BS). To solve the problem that combines continuous variables and the decoding order, we propose a deep reinforcement learning (DRL) framework that integrates a graph-based searching strategy into the deep deterministic policy gradient (DDPG) algorithm. The simulation results prove the convergence in the training of the proposed framework and its performance in various scenarios. Further-more, the results also open up an attractive research aspect when considering the sum rate maximization problem in RSMA-enhanced MIMO systems.