UAV 기반 무선전력전송 네트워크에서 강화학습을 활용한 IoT 네트워크 활성화 알고리즘

초록

Optimizing the battery use of Internet of Things (IoT) devices to reduce energy waste and maximize the lifespan of devices is one of the important research topics in IoT networks. In this study, we present an algorithm that efficiently activates deactivated IoT terminals in wireless power transmission IoT networks based on simple wireless information and power transfer (SWIPT) radio frequency (RF) communication technology supported by Unmanned Aerial Vehicle (UAV). Based on SWIPT RF communication technology, wireless power transmission UAV is used to charge IoT, and reinforcement learning that optimizes UAV's hovering point and flight path was designed to build a low-power system that minimizes the power used by UAVs that serves as a charging role. By applying the Epsilon Decay policy and the Q-learning algorithm utilizing Replay Memory technology, we present an algorithm that finally determines the flight path of UAV by moving between clusters. As a result of the simulation, it can be confirmed that the agent has been learned effectively.

키워드