CONNECT researchers throughout the centre are working on designing and building sustainable sensors and IoT systems. Here is a paper that looks to AI to reduce the energy consumption of these sensors.
Timely and sustainable: Utilising correlation in status updates of battery-powered and energy-harvesting sensors using Deep Reinforcement Learning
Authors: Jernej Hribar, Luiz A. DaSilva, Sheng Zhou, Zhiyuan Jiang, Ivana Dusparic
Journal: Computer Communications
What’s this paper about?
This paper explores the use of Age of Information (AoI) as a metric to evaluate the freshness of information gathered by various sources, such as sensors. AoI measures the time that has elapsed since the latest update was generated by the sensor and has arrived at the receiver. AoI provides a different way of looking at data collection in future networks, and our study utilizes it to create an energy-efficient method for collecting data from energy-limited sensors, including those powered by non-rechargeable batteries or energy harvesting techniques.
What have you discovered?
We have discovered that using Deep Reinforcement Learning (DRL), a category of machine learning and artificial intelligence where intelligent machines can learn from their actions similar to the way humans learn from experience, it is possible to design a scheduling mechanism capable of minimising sensors’ energy consumption. The scheduler’s objective is to balance the accuracy and timeliness of the information by deciding how often a sensor should transmit its information. By utilizing DRL, we developed a solution that adapts the sensors’ updating strategy considering factors such as available energy, current update interval, and information freshness as measured through AoI
This is a unique solution to enhance the energy efficiency of sensors as it takes into account the correlation between readings, timeliness of information, and the energy capabilities of each sensor. Employing our solution can help us develop more sustainable sensor deployments in the future.
Read the full paper here (pdf).
CONNECT is the world leading Science Foundation Ireland Research Centre for Future Networks and Communications. CONNECT is funded under the Science Foundation Ireland Research Centres Programme and is co-funded under the European Regional Development Fund. We engage with over 35 companies including large multinationals, SMEs and start-ups. CONNECT brings together world-class expertise from ten Irish academic institutes to create a one-stop-shop for telecommunications research, development and innovation.
Homepage FeaturePaper HighlightResearch Highlights