Reinforcement Learning Based Technique for Interference Management in UAV Aided HetNets

In this paper, we consider the sum rate maximization problem in a downlink unmanned aerial vehicles (UAV) assisted heterogeneous networks (HetNets) to mitigate existing interference. We propose to manage the interference by jointly optimizing the position of the UAV as well as power allocation. The proposed optimization problem is non-convex due to the non-linearity in the objective function and the constraints. We propose a near-optimal solution that breaks down the problem into two consecutive problems. First, we proposed a reinforcement learning-based technique to solve the UAV positioning problem. Then, a particle swarm optimization (PSO) based technique is applied to the power allocation problem. Simulation results demonstrate the efficiency of the proposed algorithm.