Site Search
Computer Science

Photos

Alioune Ngom, Ph.D.Dr. Alioune Ngom
Dr. Alioune Ngom
Dr. Scott GoodwinDr. Scott Goodwin
Dr. Scott Goodwin
Lambton TowerLambton Tower
Lambton Tower
Dr. Robert KentDr. Robert Kent
Dr. Robert Kent
Christie Ezeife, Ph.D.Dr. Christie Ezeife
Dr. Christie Ezeife
Robin Gras, Ph.D.Dr. Robin Gras
Dr. Robin Gras
Xiaobu Yuan, Ph.D.Dr. Xiaobu Yuan
Dr. Xiaobu Yuan
Dr. Luis RuedaDr. Luis Rueda
Dr. Luis Rueda
Jessica Chen, Ph.D.Dr. Jessica Chen
Dr. Jessica Chen
Arunita Jaekel, Ph.D.Dr. Arunita Jaekel
Dr. Arunita Jaekel
Imran Ahmad, Ph.D.Dr. Imran Ahmad
Dr. Imran Ahmad
Dr. Ziad Kobti lecturingDr. Ziad Kobti
Dr. Ziad Kobti
Windsor WaterfrontWindsor Waterfront Park
Windsor Waterfront Park

Passive RFID Rotation Dimension Reduction Via Aggregation

Add this event into your calendar using the iCAL format
  • Fri, 05/05/2017 - 10:00am - 12:00pm




Passive RFID Rotation Dimension Reduction Via Aggregation

MSc Thesis Defense by:

Eric Matthews

Date:  Friday, May 5, 2017
Time:  10:00 am – 12:00 pm
Location: Essex Hall, 122

Abstract: Radio Frequency IDentification (RFID) has applications in object identification, position, and orientation tracking. RFID technology can be applied in hospitals for patient and equipment tracking, stores and warehouses for product tracking, robots for self-localisation, tracking hazardous materials, or locating any other desired object. Efficient and accurate algorithms that perform localisation are required to extract meaningful data beyond simple identification. A Received Signal Strength Indicator (RSSI) is the strength of a received radio frequency signal used to localise passive and active RFID tags. Many factors affect RSSI such as reflections, tag rotation in 3D space, and obstacles blocking line-of-sight. LANDMARC is a statistical method for estimating tag location based on a target tag’s similarity to surrounding reference tags. LANDMARC does not take into account the rotation of the target tag. By either aggregating multiple reference tag positions at various rotations, or by determining a rotation value for a newly read tag, we can perform an expected value calculation based on a comparison to the k-most similar training samples via an algorithm called K-Nearest Neighbours (KNN) more accurately. By choosing the average as the aggregation function, we improve the relative accuracy of single-rotation LANDMARC localisation by 10%, and any-rotation localisation by 20%.

Thesis Committee:
Internal Reader: Dr. Robert Kent
External Reader: Dr. Wladyslaw Kedzierski
Advisor: Dr. Richard Frost
Co-Advisor: Dr. Roman Maev
Chair: Dr. Jessica Chen



csgradinfo@uwindsor.ca
(519)253-3000


See More: