Wireless sensor network implementations on a testbed platform

Wireless sensor network implementations are developed for three different problems: data efficient quickest change detection, energy efficient sleep control for application to the tracking problem, and diffusing point source localization and tracking. For each of these three problems work in the field is briefly examined before simulations and implementations are presented. Particular emphasis is placed on the tracking problem where a small extension is made to previous work in the field. A sub-optimal algorithm is developed that is computationally feasible for implementation purposes. This solution is compared to the more optimal solution already derived, and is then implemented on a testbed network of wireless sensors. Results from the testbed are compared, and shown to be superior in performance, to duty cycling, which is the current standard for power efficiency in a sensor network. The implementations of the data efficient quickest change detection and the energy efficient sleep control for tracking algorithms both work well, and encourage immediate development for applied use. The implementation for the diffusing point source localization and tracking algorithm is by comparison less ready for deployment, and requires further development before applied use can be seriously considered.

Name thesis:degree_name

M.S.

Level thesis:degree_level

Thesis

Discipline thesis:degree_discipline

Electrical & Computer Engr

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2013

Author dc:creator

• Harris, Kyle

Contributors dc:contributor

• Veeravalli, Venugopal V.

• wireless sensor network
• testbed
• tracking network
• quickest change detection
• partially observable Markov decision process (POMDP)
• energy-efficiency

Statement dc:rights

• Copyright 2013 Kyle A. Harris

Language dc:language

en

Handle dc:identifier

http://hdl.handle.net/2142/44181

OAI identifier oai:identifier

oai:www.ideals.illinois.edu:2142/44181