Correlated Status Update of Energy Harvesting Sensors Based on Reinforcement Learning

With the development of the Internet of Things, more and more sensors are deployed to monitor the environment status. Due to the limitation on the deployment cost, a large number of sensors need to be deployed wirelessly without stable grid power supply. Therefore, on the one hand, the wireless sensor needs to save energy as much as possible to extend its working time. On the other hand, it needs to sense and transmit timely and accurate information for real-time monitoring. In this paper, we propose a status update control mechanism where an edge node forwards status information from energy-limited energy harvesting sensors to end users based on reinforcement learning. By interacting with the environment, the edge node learns to use the correlation between state updates, which can reduce sensors' energy consumption while ensuring users' quality of service. Simulation results show that compared with several baseline methods, the proposed mechanism can well balance the energy cost and the information error, and significantly reduces the update frequency while ensuring the accuracy of information.