A Class Incremental Learning Method With Forward-Compatible and Covariance-Aware for Specific Emitter Identification
IEEE Communications Letters · 2025-09 · 29 (9) · 2153-2157
DOI: 10.1109/lcomm.2025.3588238 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Communications Letters, 2025-09. Verfasst von Xiaoyu Shen und 5 weiteren Autor:innen. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 11 Literaturangaben. Gefördert von: Project of Natural Science Foundation of China, Outstanding Youth Scientific Research Project of Universities in Anhui Province.
Abstract (Originalsprache)
Specific Emitter Identification (SEI) is essential for IoT security. Due to the continuous emergence of new communication devices in the real world, SEI needs to cope with an increasing number of transmitter categories. A trained recognition model needs to possess the capability to continuously learn new devices. This letter proposes a novel class incremental learning method based on forward compatibility and covariance awareness, named FCCA. Specifically, this letter devises a virtual signal class generation approach and an integrated loss function to expand the feature space for incremental categories while preserving valid feature representations. During the incremental phase, FCCA uses a frozen feature extractor to obtain category feature embeddings and models feature covariance relationships, helping the classifier better differentiate between categories. Experimental results on benchmark datasets demonstrate that FCCA outperforms other methods. It also demonstrates excellent performance on few-shot class incremental problems.
Förderung
- Project of Natural Science Foundation of China
- Outstanding Youth Scientific Research Project of Universities in Anhui Province
Zitieren
APA
Shen, X., Zhang, J., Qiao, X., Shang, Z., Wang, M., & Zhang, T. (2025). A Class Incremental Learning Method With Forward-Compatible and Covariance-Aware for Specific Emitter Identification. IEEE Communications Letters, 29(9), 2153-2157. https://doi.org/10.1109/lcomm.2025.3588238
Vancouver
Shen X, Zhang J, Qiao X, Shang Z, Wang M, Zhang T. A Class Incremental Learning Method With Forward-Compatible and Covariance-Aware for Specific Emitter Identification. IEEE Communications Letters. 2025;29(9):2153-2157. doi:10.1109/lcomm.2025.3588238
BibTeX
@article{Shen2025,
author = {Shen, Xiaoyu and Zhang, Jiang and Qiao, Xiaoqiang and Shang, Zhihui and Wang, Min and Zhang, Tao},
title = {A Class Incremental Learning Method With Forward-Compatible and Covariance-Aware for Specific Emitter Identification},
journal = {IEEE Communications Letters},
year = {2025},
volume = {29},
number = {9},
pages = {2153--2157},
doi = {10.1109/lcomm.2025.3588238}
}
EUVIMED