Through a UWB Looking Glass: Data Reduction Strategies for a Greener Planet

Ashutosh Dhekne, Ph.D., Assistant Professor, School of Computer Science, Georgia Tech

February 15, 2024, 3:15 - 4:15 PM ET
Hybrid Event - Teams Link

BBISS Offices, 760 Spring Street, Suite 118
Refreshments will be served.

Abstract: Cameras serve as an important data source for many IoT applications. However, cameras produce a very large amount of data that wastes resources for storage, transport, and processing. We explore an alternative where ultra-wideband wireless radios measure distance between two devices allowing us to capture a miniscule amount of data compared to video traffic. We show how the movements of a pen on a whiteboard can be captured using UWB to create real-time videos of an instructor writing on the whiteboard at a fraction of the data cost. We also show an application that helps maintain distancing at public places without generating enormous amount of contact tracing data. Similarly we explore how body signals and hand gestures can be digitized using a set of small wearable devices on the signaler’s body measuring distances between the body. We will then touch upon a replacement for security cameras in homes by instead relying on the changes in wireless signal signatures when intrusion occurs. Overall, this talk will look at the sensing world from a UWB lens instead of a camera lens.

Bio: Ashutosh Dhekne is an assistant professor in the School of Computer Science at Georgia Tech. He received his Ph.D. from the University of Illinois at Urbana Champaign and masters from IIT Bombay. He is a recipient of the NSF Career award. His research interests are in mobile computing, IoT, and wireless networking.