A Q-learning based Communication Resource Allocation Scheme for Unmanned Aerial Vehicles to Single Antenna Users

To improve the maximum and minimum user rates of single antenna communication and achieve fair allocation of network resources, a reinforcement learning based resource allocation scheme was proposed, where UAV provides auxiliary communication. Due to the fact that the mathematical model of this scheme is a non convex optimization problem that is difficult to directly solve, the scheme is divided into two stages: first, the path planning stage of the UAV, using the Q-Learning algorithm to solve the problem, and improving the performance of the Q-Learning algorithm by using genetic algorithm; Then, based on this, resource allocation is carried out, and power is allocated to users by using the water filling algorithm according to their channel status. The simulation results show that the algorithm has good throughput and fairness.