Power Control for Anti-Jamming Communications in Large-Scale UAV Networks: A Mean Field Game

This paper investigates an energy-efficient commu-nication problem in large-scale UAV base stations (UAV-BSs) networks by mean field game (MFG) approach. Considering the complicated internal interference and external dynamic jamming, the cost function of UAV-BSs is first designed. Through adjusting the anti-jamming power control policy, the cost is minimized to raise resource utilization efficiency under the quality of service requirements. Then, the optimization problem is formulated as a differential game, where the dynamics of jamming and the energy consumption are considered into the players' state. To achieve low-complexity optimization for large-scale networks, we introduce mean field and propose an adaptive MFG approach, where the mean field varies with the dynamics of the jamming. A distributed algorithm for anti-jamming power control policy is designed to obtain the Nash Equilibriums. Simulation results show that compared with state-of-the-art approaches, the proposed adaptive MFG improves the resource utilization efficiency by at least 25%, which indicates the effectiveness of the proposed approach in large-scale networks with dynamic jamming.