How to Deploy Multiple UAVs for Providing Communication Service in an Unknown Region?

Abstract

It is an effective way to deploy unmanned aerial vehicle (UAV)-mounted base stations for keeping ground communication in an unknown region when the ground communication facility is unavailable for some reason. This letter investigates the joint placement and power allocation problem of multiple UAVs for ground terminals’ (GTs) communication in an unknown region. In the beginning, UAVs have no information of GTs, which results in that UAVs have to keep flying for information gathering until find the optimal locations. In practice, however, imperfect channel state information (CSI) is obtained by UAVs and only neighboring UAVs can communicate with each other. To overcome these challenges, we leverage game theory to model this problem and propose a robust and distributed learning algorithm to converge to a stochastic stable state of maximizing the optimization objective, i.e., the sum of $\alpha $ -fairness of all GTs.