Date of Graduation

5-2025

Document Type

Thesis

Degree Name

Master of Science in Computer Science (MS)

Degree Level

Graduate

Department

Electrical Engineering and Computer Science

Advisor/Mentor

Pan, Yanjun

Committee Member

Wu, Jingxian

Second Committee Member

Zhang, Lu

Keywords

Neural networks; Radio frequency fingerprinting; Wireless security

Abstract

Radio frequency (RF) fingerprints, caused by unique imperfections in communication hardware, offer a promising solution for zero-trust security. However, existing RF fingerprinting techniques, which aim to extract these signatures from transmitters to uniquely identify devices, often struggle with robustness in the face of temporal and spatial variations in real-world, time-varying wireless environments. For example, a neural network trained on RF signals collected on Day 1 can experience a significant performance drop when tested with data from Day 2.

To address this challenge, we propose a novel, robust RF fingerprinting method based on Physics-Informed Neural Networks (PINNs). Rather than training the model solely on the received signal, a complex mixture of RF fingerprints, time-varying channel conditions, and random channel noise, we incorporate invariant radio physics, specifically the estimated Carrier Frequency Offset (CFO), to guide the model’s learning. In addition, we input a distilled, frequency-domain equalized signal to mitigate the effects of dynamic wireless propagation conditions. Extensive experiments using USRPs were conducted to collect real-world RF data, and the proposed PINN model was benchmarked against state-of-the-art approaches. While all three methods achieved high classification accuracy (>98%) when training and testing on the same day, the cross-day performance of baseline models dropped to approximately 35%. In contrast, the proposed PINN model maintained an accuracy of 97.54% across days. These results demonstrate that embedding radio physics into the learning process significantly enhances model robustness, offering a more resilient and practical path for deploying RF fingerprinting in real-world security systems.

Download

Share

COinS