Title: Autonomous Moisture Continuum Sensing Network: Intelligent and Energy Efficient in situ Wireless Sensor Networks
Presenting Author: Ruzbeh Akbar
Organization: Massachusetts Institute of Technology
Co-Author(s): Dara Entekhabi (PI), MIT; Mahta Moghaddam (Co-I), University of Southern California; Agnelo R. Silva, METER Group; Ruzbeh Akbar, MIT, PostDoc

Abstract:
Environmental in situ sensor networks, e.g., SCAN, AmeriFlux, USCRN, etc., are key components in long-term monitoring and observation of Earth's water, carbon, and energy cycles. Energy requirements to maintain network operations are typically satisfied by use of solar panels and large capacity rechargeable batteries. Additionally, existing in situ networks operate on a fixed and predetermined sampling schedule, e.g. 20 min or hourly. In this project, we build upon existing WSN infrastructure and prior developments from the Soil Moisture Sensing Controller and Optimal Estimator (SoilSCAPE) (TRL ~7). Specifically, we will present new technology concepts and developments towards enabling a WSN to intelligently and autonomously decide when and where to collect in situ soil moisture and canopy sap-flow measurements while simultaneously minimizing overall energy consumption. Two technology features will be discussed. First includes new concepts in wireless communication, especially Wakeup-on-Radio (WoR, TRLin = 2) transceiver modules, and their role in in situ WSNs and on-demand, or event-driven, operations. The second feature includes machine learning based WSN sampling, scheduling and decision-making strategies (TRLin = 2) to minimize energy consumption, while maximum overall measurement information content.