Sum-Rate Maximization for RSMA-Enabled Energy Harvesting Aerial Networks With Reinforcement Learning

In this paper, we propose a joint power and precoder design framework for energy-harvesting aerial networks, where an aerial base station serves multiple users via rate-splitting multiple access (RSMA) using harvested energy. To maximize the sum-rate from the long-term perspective, we utilize a deep reinforcement learning (DRL) approach to allocate optimal transmission power at each time based on the randomness property of the channel environment, harvested energy, and battery power information. Moreover, we employ sequential least squares programming (SLSQP) to design the RSMA precoder maximizing the sum-rate with the allocated power. Numerical results show the superiority of the proposed scheme over baseline methods in terms of the average sum-rate performance.