Event Details
Energy Efficient Data Collection From Wirelessly Powered IoT Sensors: Session-Specific Analysis and Optimal Design with Deep Reinforcement Learning
Presenter: Fang Xu
Supervisor:
Date: Wed, June 8, 2022
Time: 14:00:00 - 15:00:00
Place: via Zoom - please see link below
ABSTRACT
Zoom link: https://uvic.zoom.us/j/8565167653?pwd=L0NHMmxrU2RNSDQrZWZUM29hMy92dz09
Meeting ID: 856 516 7653
Password: 113773
Abstract:
Reliable and energy efficient data collection from resource-limited sensors is very critical to the success of future Internet of Things (IoT). We investigate the problem of energy consumption minimization during data collection from a wirelessly-powered sensor. Specifically, we determine the optimal data collection parameters, in terms of charging duration and charging power as well as sensor transmission rate, in real time according to the instantaneous channel condition while satisfying a certain latency constraint. For ideal rate adaptive transmission with linear energy harvesting, we derive closed-form expressions for all optimal transmission parameters. We also establish the condition on channel quality for successful data collection under a latency constraint. For more practical finite block-length transmission with non-linear energy harvesting, we develop a deep reinforcement learning (DRL) solution for determining near optimal transmission parameters in real time during the online operation. We also propose an online tuning scheme to cater for model inaccuracy and environment variation. The accuracy and effectiveness of our proposed approaches has been verified by comparing with benchmark schemes, i.e. gradient ascent and exhaustive search. Our DRL-based approach has broad applicability and can solve other real-time optimal design problems in wireless communications.