Concealing Radio Frequency Fingerprints via Active Adversarial Perturbation
Bibliographic record
Abstract
Radio frequency fingerprinting (RFF) is an emerging technology to identify a radio device via recognizing its unique feature that originated from manufacturing imperfections. Nevertheless, these invariant hardware characteristics can be exploited by adversaries, potentially compromising device privacy, even with user anonymity measures in place. This paper addresses the challenge of concealing RF fingerprints in the context of pilot signal-based fingerprint identification. Specifically, artificial perturbations are applied to the radio signal, focusing on the pilot signal during radio transmission. To prevent neural models from recognizing the unique hardware-induced features embedded in the pilot signal, we generate a set of active artificial perturbations through adversarial attack optimization. The perturbed signal aims to mislead the neural models in device identification, thereby obfuscating RF fingerprints and preventing unauthorized identification from potential attackers. To ensure normal communication is unaffected, we theoretically analyze the channel estimation error caused by the perturbation to the pilot signal, and show that the impairments to communication can be controlled within a limited range. To demonstrate effectiveness, we implement the entire fingerprinting and de-fingerprinting process based on a 4G LTE testbed. Extensive experiments demonstrate that the proposed method can effectively conceal the device-specific feature against RF fingerprinting neural models.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.005 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".