Publication Date

2020

Document Type

Thesis

Committee Members

Michael A. Saville, Ph.D., P.E. (Committee Chair); Saiyu Ren, Ph.D. (Committee Member); Henry Chen, Ph.D. (Committee Member); Joshua N. Ash, Ph.D. (Committee Member)

Degree Name

Master of Science in Electrical Engineering (MSEE)

Abstract

Radio Frequency Fingerprinting (RFF) research typically uses expensive, laboratory grade receivers which have high dynamic range, very stable oscillators, large instantaneous bandwidth, multi-rate sampling, etc. In this study, the RFF effectiveness of lower grade receivers is considered. Using software-defined radios (SDRs) of different cost and performance, a linear regression model is developed to predict RFF performance. Unlike two previous studies of SDR effectiveness that used commercial and lab-grade SDRs, the experiment here focused on hobbyist and commercial-grade SDRs (RTL-SDR, B200-mini, N210). A regression model is proposed for a generic SDR. Using a full-factorial experiment matrix, the gain, sample rate, and signal-to-noise ratio (SNR) were selected as the common control factors. The transmit sources were three commercially-available, general purpose, wireless transmitters of the same model. An SDR performance index (SPI) was developed from the percent correct classification using the Random Forest classifier for each SDR and for a generic SDR. The RFF results show that the lower-cost SDRs record the data with enough fidelity to achieve over 90% classification accuracy.

Page Count

61

Year Degree Awarded

2020

Creative Commons License

Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 License.

ORCID ID

0000-0003-3603-1311


Share

COinS