UAV-Assisted Emergency Communications: An Extended Multi-Armed Bandit Perspective

In this letter, we investigate how a rotary-wing unmanned-aerial vehicle (UAV) acts as a wireless base station to provide emergency communication service for a post-disaster area with unknown user distribution. The formulated optimization task is to find out the optimal path starting and ending at the same point to serve as many as possible users under limited battery capacity. We show that this problem can be transformed to an extended multi-armed bandit (MAB) problem, for which we propose two efficient path planning algorithms. Simulation results show that, in terms of the total number of served users, our proposed algorithms outperform the straightforward helical path that scans the entire area by circles with increasing radius.