EUVIMEDEUVIMED entdecken
Crossref

A Multi-Physics Drift Model for Robust Radio Frequency Fingerprinting: From Theory to Application

Zichuan Yu, Hongyu Ge, Xusheng Tang, Hongyuan Liu

IEEE Communications Letters · 2026 · 30 · 2530-2534

DOI: 10.1109/lcomm.2026.3710940 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Communications Letters, 2026. Verfasst von Zichuan Yu und 3 weiteren Autor:innen. Die Arbeit nennt 19 Literaturangaben. Gefördert von: Open Fund of the Key Laboratory of Short-Range Radio Equipment Testing and Evaluation of the Ministry of Industry and Information Technology.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Radio Frequency Fingerprint Identification (RFFI) is severely bottlenecked by physical fingerprint drift. Moving beyond treating this drift as a black-box statistical shift, we propose a comprehensive multi-physics theoretical model that rigorously decouples macro-scale thermodynamics, meso-scale stress relaxation, and micro-scale circuit modulation. As a direct application and empirical validation of this theory, we introduce the Universal Physical Drift Data Augmentation (Uni-PDA) framework. Operating purely via target-free tensor transformations, Uni-PDA injects deterministic physical drift boundaries into a single steady-state dataset. Extensive experiments across diverse architectures demonstrate that this theory-guided synthesis improves cross-drift generalization without target-domain data or additional inference-stage overhead. When integrated with Decoupling-RFFI, Uni-PDA achieves 88.8% drift-domain accuracy while maintaining 93.9% steady-state accuracy, reducing the absolute accuracy drop from 14.1 to 5.1 percentage points.

Förderung

  • Open Fund of the Key Laboratory of Short-Range Radio Equipment Testing and Evaluation of the Ministry of Industry and Information Technology

Zitieren

APA

Yu, Z., Ge, H., Tang, X., & Liu, H. (2026). A Multi-Physics Drift Model for Robust Radio Frequency Fingerprinting: From Theory to Application. IEEE Communications Letters, 30, 2530-2534. https://doi.org/10.1109/lcomm.2026.3710940

Vancouver

Yu Z, Ge H, Tang X, Liu H. A Multi-Physics Drift Model for Robust Radio Frequency Fingerprinting: From Theory to Application. IEEE Communications Letters. 2026;30:2530-2534. doi:10.1109/lcomm.2026.3710940

BibTeX
@article{Yu2026,
  author = {Yu, Zichuan and Ge, Hongyu and Tang, Xusheng and Liu, Hongyuan},
  title = {A Multi-Physics Drift Model for Robust Radio Frequency Fingerprinting: From Theory to Application},
  journal = {IEEE Communications Letters},
  year = {2026},
  volume = {30},
  pages = {2530--2534},
  doi = {10.1109/lcomm.2026.3710940}
}
Kostenlos starten →