Mode Selection in UAV-aided Vehicular Network: an Evolutionary Game Approach

Abstract

In vehicular networks, safety-related applications require ultra high transmission reliability. An unmanned aerial vehicle (UAV) aided vehicular network is explored to enhance the reliability, where UAVs provide additional communication resources for vehicles. In this scenario, three communication modes are available for vehicles, i.e. vehicle to base station V2B, vehicle to vehicle V2V, and vehicle to UAV V2U. The transmission reliability highly depends on the selection of communication modes under different channel conditions and resource constraints, and thus the key issue is to appropriately select optimal communication modes to achieve the best reliability. In this paper, a mode selection approach based on evolutionary game is proposed in UAV-aided vehicular network, and an evolutionarily stable strategy is obtained. With the proposed algorithm, the proportion of vehicles in different communication modes can converge quickly to the evolutionary equilibrium where no vehicle has an incentive to change its communication mode. Numerical results show that our proposed algorithm exhibits a fast and controllable convergence and achieves higher transmission reliability with lower cost of resource utilization, compared with the selfish and random selection schemes.