Total: 1
The development of wireless power transfer (WPT) and Internet of Things (IoT) offers significant potential but faces challenges such as limited energy supply, dynamic environmental changes, and unstable transmission links. This paper presents an unmanned aerial vehicle (UAV)-assisted data collection and WPT scheme to support batteryless sensor (BLS) networks in remote areas. In this system, BLSs harvest energy from the UAV and utilize the harvested energy to transmit the collected data back to the UAV. The goal is to maximize the collected data volume and fairness index while minimizing the UAV energy consumption. To achieve these objectives, an optimization problem is formulated to jointly optimize the transmit power and UAV trajectory. Due to the non-convexity and dynamic nature of the problem, a deep reinforcement learning (DRL)-based algorithm is proposed to solve the problem. Specifically, this algorithm integrates prioritized experience replay and the performer module to enhance system stability and accelerate convergence. Simulation results demonstrate that the proposed approach consistently outperforms benchmark schemes in terms of collected data volume, fairness, and UAV energy consumption.