Delay Optimization in Multi-UAV Edge Caching Networks: A Robust Mean Field Game

Abstract

The data requirements of a large number of users incur significant delay to base stations/small base stations (BSs/SBSs) in edge caching networks (ECNs). Specially, in certain special scenarios, such as sports events, parades and nature disasters, it is a great challenge to deploy fixed BSs/SBSs. Thus, a kind of flexible replacement of BSs/SBSs, unmanned aerial vehicles (UAVs), are considered as the caching platform to serve users. Hence, this paper proposes a distributed delay optimization algorithm by using the massive UAVs to cache the request contents of users. The purpose of each UAV is to facilitate BSs/SBSs to minimize user delay in downloading content based on the popularity of content and the flight strategy. In addition, we consider the main disturbance caused by the atmospheric turbulence. For the control of large-scale UAVs, we take the robust mean field game theory to model this kind of caching and dynamic flight strategy problem, in which the atmospheric turbulence is described by the disturbance term in the drift function. The simulation results show that the proposed algorithm can reduce the download delay under the limited energy consumption and finally achieve the equilibrium.