Deep learning has been widely used in radio frequency (RF) fingerprinting. Despite its excellent performance, most existing methods only consider a closed-set assumption, which cannot effectively tackle signals emitted from those unknown devices that have never been seen during training. In this letter, we exploit prototype learning for open-set RF fingerprinting and propose two improvements, including consistency-based regularization and online label smoothing, which aim to learn a more robust feature space. Experimental results on a real-world RF dataset demonstrate that our proposed measures can significantly improve prototype learning to achieve promising open-set recognition performance for RF fingerprinting.
@article{arxiv.2306.13895,
title = {Open-Set RF Fingerprinting via Improved Prototype Learning},
author = {Weidong Wang and Hongshu Liao and Lu Gan},
journal= {arXiv preprint arXiv:2306.13895},
year = {2023}
}