CRII: NeTS: RUI: Fundamentals of Wireless Networking using Battery-Free Devices in Inhomogeneous Environments

University Of Nebraska-Lincoln, Lincoln NE

Investigators

Abstract

Precision agriculture is a transformative technology that consumes fewer resources and generates more yields by using recently developed technologies such as wireless sensor networks and autonomous robots. Wireless underground sensors that are buried in the soil to provide rich information about soil nutrition will play an important role in precision agriculture. Although they can significantly improve farm yields, wireless underground sensors are costly and hard to maintain since they are deployed in the inaccessible soil medium, which may reduce farmers' profits. This project is expected to radically change the wireless underground sensor network by significantly reducing its cost and complexity. The idea of employing low-cost battery-free wireless sensors with magnetic induction backscatter communications is transformative since these sensors do not require battery replacement and consume negligible energy. The proposed research contributes to a much wider vision of providing ubiquitous connectivity for in-situ sensors in inhomogeneous media such as concrete walls, underwater, and intra-body, which can enable a large number of civilian and military applications. This project also includes a plan to create various educational and research experiences for underrepresented minority students to promote diversity in the workforce. This project aims to develop adaptable and scalable wireless underground sensor networks using low-cost battery-free sensors with magnetic induction backscatter communications. The ultimate goal is to replace today’s bulky battery-powered wireless sensors with battery-free ones for in-situ sensing applications in inhomogeneous environments. Since the inhomogeneous media destroy existing wireless communication and networking solutions that are based on geometric principles, new theories and algorithms will be developed to overcome this problem and advance our knowledge of wireless network design in complex inhomogeneous environments. This project will develop physical layer solutions for magnetic induction backscatter communications, including channel estimation, blind beamforming, and energy harvesting. A reconfigurable testbed will be built to verify the proposed solutions, upon which channel measurement campaigns will be performed to derive a comprehensive channel model. Soil dynamics can dramatically change network connectivity and throughput. To this end, adaptive network control algorithms based on machine learning will be used to provide ubiquitous coverage, maintain network integrity, and prolong the system’s lifetime. This project will also develop localization algorithms using magnetic induction backscatter communications to track the locations of invisible underground sensors. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.