Age of Information Minimization for Buffer-Aided UAV Wireless Communications

The utilization of data buffer in unmanned aerial vehicle (UAV) introduces a dual role in the age of information (AoI) performance. Despite the enhancement of data delivery quality resulting from flexible and efficient data transmission, buffering may also risk increasing AoI by causing data aging. Against this background, this paper investigates the AoI minimization problem in buffer-aided UAV wireless communications while incorporating the effects of UAV buffering dynamics and limited buffer size. Specifically, we formulate the problem as a partially observable Markov decision process (POMDP) and propose a deep recurrent Q-network (DRQN) -based algorithm to jointly optimize UAV trajectory planning and buffering decisions. Simulation results exhibit the superiority of the proposed algorithm in terms of reducing the average AoI. Moreover, the study sheds light on the impact of UAV buffer size on AoI optimization, yielding essential design for buffer-aided UAV communications.