UAV Position Optimization Based on Information Freshness: A Mean Field Game Approach

In order to deal with the challenge of data traffic surge, unmanned aerial vehicle (UAV) is widely investigated to provide timely internet services for ground users due to its mobility and flexibility. Meanwhile, for the purpose of measuring the timeliness, reliability and security of user's data updates, the freshness of information has also received extensive attention. In this paper, we formulate an ultra-dense network of multi-UAVs assisted base stations (BSs) serving users, and minimize the age of information (AoI) of the system by optimizing the positions of the UAVs to improve information freshness. To simplify the complicated large dense network model, a mean field game (MFG) method is proposed to jointly optimize the positions of these UAVs, taking into account the interference between UAVs and the flight energy consumption. In addition, a deep deterministic policy gradient (DDPG) algorithm is implemented to obtain the mean field equilibrium solution. The simulation results verify that the proposed algorithm can effectively reduce the AoI of the system.