Air-ground integrated assisted proactive eavesdropping

Benefiting from the rapid development of unmanned aerial vehicle (UAV) technology, UAVs have also received extensive attention in the field of communication. In this letter, we investigate an air-ground proactive eavesdropping system in which a legitimate ground eavesdropper can actively eavesdrop on suspected ground communication links with the assistance of a UAV. To improve the eavesdropping performance of the system, the optimal trajectory of the UAV and the appropriate power allocation ratio are sought to maximize the eavesdropping rate. A Double-Dueling DQN (D3QN) based scheme for maximizing the eavesdropping rate is proposed through deep reinforcement learning. The joint optimization of UAV trajectory and power allocation ratio is achieved using the D3QN algorithm. From the numerical results, the optimization scheme can improve the eavesdropping rate of the system.