Wireless sensing: Material identification and localization

Dhekne, Ashutosh Makrand

Permalink

https://hdl.handle.net/2142/105676 Copy

Description

Title

Wireless sensing: Material identification and localization

Author(s)

Dhekne, Ashutosh Makrand

Issue Date

2019-07-12

Director of Research (if dissertation) or Advisor (if thesis)

Roy Choudhury, Romit

Doctoral Committee Chair(s)

Roy Choudhury, Romit

Committee Member(s)

• Nahrstedt, Klara
• Godfrey, Philip B
• Sundaresan, Karthikeyan
• Lane, Nicholas

Department of Study

Computer Science

Discipline

Computer Science

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Keyword(s)

Wireless Sensing, Localization, Material Identification, UWB, ultra-wideband, Liquid Identification, Sports Analytics, IoT for Sports, Mobile Computing, Drones, UAV,

Abstract

Wireless signals are everywhere around us, and they have truly revolutionized the world by all standards. When one thinks of this revolution, one envisions the advances in wireless communication—TV broadcasts, FM radios, WiFi, Bluetooth, cellular mobile phones, and even wireless chips inside the human body. What gets less appreciated, however, is that wireless signals can also be a powerful sensor. The fact that wireless signals touch and penetrate all objects in our environment, and bounce back, make them a powerful lens to view our world through. This thesis focuses on using wireless signals as sensors. We will explore how modifications to wireless signal propagation can reveal the physical properties of the materials that these signals have passed through. This enables identification of materials without touching them or performing any chemical analysis on them. We will show the ability to distinguish between closely related liquids, such as Pepsi and Coca-Cola, or distilled water and mineral water, by simply passing wireless signals through the liquids, and analyzing the signals that emerge on the other side. The propagation delay of wireless signals when passing through air can reveal the distance between a transmitter and a receiver. We show how this primitive can be extended for localization with applications to sports, battlefields, and emergency response. Through modifications to the distance measurement mechanisms, we show how localization is possible even when wireless devices are constantly under motion. We end by discussing future directions in which both of these sensing techniques can be extended. Under the right conditions, it might be possible to localize an object to 5mm precision with applications in robotic machines, augmented reality, and virtual reality. We then discuss the possibility of using reflections of wireless signals, for example, to determine soil moisture content in agricultural fields.

Graduation Semester

2019-08

Type of Resource

text

Permalink

http://hdl.handle.net/2142/105676

Copyright and License Information

2019 Ashutosh Makrand Dhekne

Owning Collections