Periodic Data Scheduling Scheme for Power Internet of Things Based on Age of Information *

In this paper, we study a real-time PIoT monitoring system in which multiple terminal nodes send periodic data to maintain the Age of Information (AoI) at the base station. The batteries of the terminal nodes can be charged through wireless energy transfer (WET). We investigate the optimal resource scheduling scheme that jointly optimizes WET and the scheduling of periodic packet transmissions. The objective is minimizing the long-term average weighted sum of AoI value at the base station. To address this optimization problem, we set it as a Markov decision process (MDP) with finite state and action spaces. The Deep Q Network (DQN) algorithm is applied to learn the age-optimal policy in a computationally-efficient manner. Simulation results demonstrate the performance of the DQN algorithm in optimizing the long-term average weighted sum of AoI value at the base station.