Adaptive Height Optimisation for Cellular-Connected UAVs using Reinforcement Learning

Providing reliable connectivity to cellular-connected UAVs can be very challenging; their performance highly depends on the nature of the surrounding environment, such as density and heights of the ground BSs. On the other hand, tall buildings might block undesired interference signals from ground BSs, thereby improving the connectivity between the UAVs and their serving BSs. To address the connectivity of UAVs in such environments, this paper proposes a RL algorithm to dynamically optimise the height of a UAV as it moves through the environment, with the goal of increasing the throughput that it experiences. The proposed solution is evaluated using experimentally-obtained measurements from two different locations in Dublin city centre, Ireland. In the first scenario, the UAV is connected to a macro-cell, while in the second scenario, the UAVs associates to different small cells in a two-tier mobile network. Results show that the proposed solution increases 6 to 41% in throughput, compared to baseline approaches.