Oct 22 (Fri) @ 11:30am: "Enabling Novel Sensing Applications with Everyday WiFi Signals," Belal Korany, ECE PhD Defense

https://ucsb.zoom.us/j/86139204455?pwd=eXJxWkg2Qk5mRWdESVVGaUZBSTJuZz09

Abstract

Recent years have witnessed a rapid growth in the number of wirelessly connected devices, making wireless signals, such as WiFi, ubiquitous. Since these wireless signals continuously interact with our surroundings (e.g., by reflecting off of them), researchers have shown interest in utilizing these signals for sensing. In this talk, I will show how WiFi signals can be used to realize different novel sensing applications.

First, I will discuss how video data can be converted into instant RF data. Specifically, given a video footage of a person engaged in some activity, I will show how we can generate/simulate the WiFi signal that would have been measured by a WiFi device if it was near the person in the video. Given there are many available video datasets, this video-to-WiFi conversion tool will be essential as it can translate video datasets to instant RF data, thus enabling training RF sensing systems, without the need to collect any RF training data.  It further allows us to develop a cross-modal video to RF person identification system, as we shall see.

Next, I will discuss how WiFi signals can detect and monitor the nocturnal seizure activity of epilepsy patients. I will present a new mathematical model for the WiFi signals that are reflected off of the body of a sleeping patient, and then show our proposed framework for processing the WiFi signals for the purpose of seizure detection.

Finally, I will discuss our proposed approach for counting a stationary (seated) crowd using WiFi signals. More specifically, we propose a novel technique that enables WiFi signals to count a stationary seated crowd, using their natural body fidgets.  We propose that the aggregate natural fidgeting and in-place motions of a seated crowd carry crucial information on the crowd count, and show how to estimate the total number of people from the aggregate crowd fidgets, by drawing similarities between this problem and queuing theory.

Overall, our proposed sensing methodologies can enable many different applications with cheap ubiquitous WiFi signals.

Bio

Belal Korany received his B.Sc. and M.S. degrees in Electrical Engineering from Cairo University, Egypt, in 2012 and 2015, respectively. He is currently a PhD candidate in the ECE department at UCSB and is advised by Professor Yasamin Mostofi. His research interests include wireless sensing, array signal processing, and wireless communications.

Hosted by: Professor Yasamin Mostofi

Submitted by: Belal Korany <belalkorany@ece.ucsb.edu>