Reinforcement Learning for Energy-Efficient User Association in UAV-Assisted Cellular Networks

In unmanned aerial vehicle (UAV)-assisted communications, there are two significant challenges that need to be addressed - optimized UAV placement and energy-efficient user association. These challenges are crucial in meeting the quality-of-service requirements of users. To overcome these challenges, a reinforcement-learning-based intelligent solution is proposed along with a reward function that associates users with UAVs in an intelligent manner. This solution aims to improve the system's sum rate performance by consuming less energy. Simulation results are presented to demonstrate the effectiveness of the proposed approach. The results indicate that the proposed approach is more energy efficient than the benchmark scheme while improving the system's sum rate.