EUVIMEDEUVIMED entdecken
Crossref

Adversarial Samples Detection Based on Feature Attribution and Contrast in Modulation Recognition

Wenyu Wang, Lei Zhu, Yuantao Gu, Yufan Chen, Xingyu Zhou, Lu Yu

IEEE Communications Letters · 2024-11 · 28 (11) · 2483-2487

DOI: 10.1109/lcomm.2024.3463949 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Communications Letters, 2024-11. Verfasst von Wenyu Wang und 5 weiteren Autor:innen. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 14 Literaturangaben. Gefördert von: National Natural Science Foundation of China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Detecting adversarial samples is crucial for maintaining the security of automatic modulation recognition (AMR) systems, as adversarial attacks could severely compromise wireless communication. To address this threat, we propose a novel adversarial samples detection method named Null Feature Attribution Abnormality (NFAA), which leverages the target model’s interpretation of feature importance to distinguish between benign and adversarial signal samples. Furthermore, we propose the NFAA-TC method, incorporating a Triple Contrast (TC) approach to mitigate noise in signal data and enhance the performance of adversarial samples detection. Experimental results validate the effectiveness of the proposed method across various adversarial attacks and under different signal-to-noise ratio (SNR) conditions.

Förderung

  • National Natural Science Foundation of China

Zitieren

APA

Wang, W., Zhu, L., Gu, Y., Chen, Y., Zhou, X., & Yu, L. (2024). Adversarial Samples Detection Based on Feature Attribution and Contrast in Modulation Recognition. IEEE Communications Letters, 28(11), 2483-2487. https://doi.org/10.1109/lcomm.2024.3463949

Vancouver

Wang W, Zhu L, Gu Y, Chen Y, Zhou X, Yu L. Adversarial Samples Detection Based on Feature Attribution and Contrast in Modulation Recognition. IEEE Communications Letters. 2024;28(11):2483-2487. doi:10.1109/lcomm.2024.3463949

BibTeX
@article{Wang2024,
  author = {Wang, Wenyu and Zhu, Lei and Gu, Yuantao and Chen, Yufan and Zhou, Xingyu and Yu, Lu},
  title = {Adversarial Samples Detection Based on Feature Attribution and Contrast in Modulation Recognition},
  journal = {IEEE Communications Letters},
  year = {2024},
  volume = {28},
  number = {11},
  pages = {2483--2487},
  doi = {10.1109/lcomm.2024.3463949}
}
Kostenlos starten →