Date of Graduation

5-2012

Document Type

Dissertation

Degree Name

Doctor of Philosophy in Engineering (PhD)

Degree Level

Graduate

Department

Computer Science & Computer Engineering

Advisor/Mentor

Thompson, Dale R.

Committee Member

Beavers, M. Gordon

Second Committee Member

Di, Jia

Third Committee Member

McCann, Roy A.

Keywords

Applied sciences; Authenthicaton; Classification; Cyber-security; Hidden markov models; Identification; Pattern recognition; Rfid; Transmitted signals

Abstract

The ability to identify and authenticate entities in cyberspace such as users, computers, cell phones, smart cards, and radio frequency identification (RFID) tags is usually accomplished by having the entity demonstrate knowledge of a secret key. When the entity is portable and physically accessible, like an RFID tag, it can be difficult to secure given the memory, processing, and economic constraints. This work proposes to use unique patterns in the transmitted signals caused by manufacturing differences to identify and authenticate a wireless device such as an RFID tag. Both manufacturer identification and tag identification are performed on a population of 300 tags from three different manufacturers. A methodology to select features for identifying signals with high accuracy is developed and applied to passive RFID tags. The classifier algorithms K-Nearest Neighbors, Parzen Windows, and Support Vector Machines are investigated. The tag's manufacturer can be identified with 99.93\% true positive rate. An individual tag is identified with 99.8\% accuracy, which is better than previously published work. Using a Hidden Markov Model with framed timing and power data, the tag manufacturer can be identified with 97.37\% accuracy and has a compact representation. An authentication system based on unique features of the signals is proposed assuming that the readers that interrogate the tags may be compromised by a malicious adversary. For RFID tags, a set of timing-only features can provide an accuracy of 97.22\%, which is better than previously published work, is easier to measure, and appears to be more stable than power features.

Download

Share

COinS