Decoding the Collisions in RFID System

MPhil Thesis Defence


Title: "Decoding the Collisions in RFID System"

By

Mr. Lei Kang


Abstract

Traditionally, collision alone is nothing but noise. This paper adopts a 
new approach; it directly decodes the collisions without exact knowledge 
of collided packets. RFID tag uses random number (RN) to contend for tag 
identification. We re-design the RN pattern to be only one or two "1" bits 
in it, so that we can decode the collided RNs and know the number of them 
by subtracting the known symbols.

Two key tasks in RFID systems, tag identification and tag number 
estimation, can be performed in a much more efficient way. The new scheme 
makes no changes to the RFID MAC and introduces no overhead when the 
traditional reader or tags are used. We extend the RN generator on USRP 
RFID tag and implement a decoding algorithm at the physical layer on USRP 
RFID reader to demonstrate the practicability of our design. The new 
approach improves the throughput by 70% in tag identification and enhances 
the counting efficiency by orders of magnitude compared to existing 
solutions.


Date:			Tuesday, 6 July 2010

Time:			1:30pm – 3:30pm

Venue:			Room 3501
 			Lifts 25/26

Committee Members:	Prof. Lionel Ni (Supervisor)
 			Dr. Lin Gu (Chairperson)
 			Dr. Lei Chen


**** ALL are Welcome ****