HidePrint: Hiding the Radio Fingerprint via Random Noise

2024-11-10

arXiv.org

AI Summary

To address the misuse of radio frequency fingerprints (RFFs) for unauthorized device tracking, this paper proposes an RFF anonymization method ensuring communication transparency. The method introduces: (1) a novel selective RFF disclosure mechanism, enabling transmitters to reveal fingerprints exclusively to authorized receivers—thereby preserving both anonymity and authentication capability; (2) a controllable Gaussian noise injection scheme at the physical layer, which perturbs transmitted signals to obfuscate RFF features—achieving degradation of mainstream image-domain RFF classifiers to random-guess performance (accuracy ≈ 1/N) when noise standard deviation ≥ 0.02; and (3) robust adversarial evaluation across diverse link configurations and RFF extraction models, demonstrating negligible communication impact (SNR reduction ≤ 0.1 dB). To the best of our knowledge, this is the first approach achieving high-robustness, low-overhead RFF privacy protection without compromising system transparency or link quality.

Technology Category

• Machine Learning: Privacy
• Computer Vision: Motion & Tracking
• Natural Language Processing: Ethics — Bias, Fairness, Transparency & Privacy

Application Category

• Security and Privacy: Tracking, profiling, and countermeasures against them
• User Modeling, Personalization and Recommendation: User privacy protection in personalized systems
• Responsible Web: Measurement, analysis, and circumvention of Web censorship

Abstract

Radio Frequency Fingerprinting (RFF) techniques allow a receiver to authenticate a transmitter by analyzing the physical layer of the radio spectrum. Although the vast majority of scientific contributions focus on improving the performance of RFF considering different parameters and scenarios, in this work, we consider RFF as an attack vector to identify and track a target device. We propose, implement, and evaluate HidePrint, a solution to prevent tracking through RFF without affecting the quality of the communication link between the transmitter and the receiver. HidePrint hides the transmitter's fingerprint against an illegitimate eavesdropper by injecting controlled noise in the transmitted signal. We evaluate our solution against state-of-the-art image-based RFF techniques considering different adversarial models, different communication links (wired and wireless), and different configurations. Our results show that the injection of a Gaussian noise pattern with a standard deviation of (at least) 0.02 prevents device fingerprinting in all the considered scenarios, thus making the performance of the identification process indistinguishable from the random guess while affecting the Signal-to-Noise Ratio (SNR) of the received signal by only 0.1 dB. Moreover, we introduce selective radio fingerprint disclosure, a new technique that allows the transmitter to disclose the radio fingerprint to only a subset of intended receivers. This technique allows the transmitter to regain anonymity, thus preventing identification and tracking while allowing authorized receivers to authenticate the transmitter without affecting the quality of the transmitted signal.

Problem

• Preventing device identification through radio frequency fingerprinting attacks
• Obscuring transmitter fingerprints without degrading communication quality
• Enabling selective fingerprint disclosure to authorized receivers only

Innovation

• Injecting controlled Gaussian noise into signals
• Hiding device fingerprints from unauthorized eavesdroppers
• Enabling selective radio fingerprint disclosure to receivers