期刊
IEEE SIGNAL PROCESSING LETTERS
卷 30, 期 -, 页码 70-74出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/LSP.2023.3241592
关键词
Receivers; Feature extraction; Adaptation models; Signal to noise ratio; Training; Signal processing algorithms; Degradation; Radio frequency fingerprint; cross-receiver; contrastive learning; subdomain adaptation
Radio frequency fingerprint (RFF) identification is a promising approach for physical layer security. However, current deep learning (DL) based schemes often suffer from performance degradation in cross-receiver scenarios. To address this issue, a cross-receiver RFF learning scheme is proposed, which utilizes unsupervised pre-training and subdomain adaptation to enhance identification performance. Experimental results demonstrate that the proposed scheme effectively mitigates the performance degradation in cross-receiver scenarios.
Radio frequency fingerprint (RFF) identification is emerging as an attractive paradigm for physical layer security. Despite the exceptional accuracy achieved by deep learning (DL) based schemes, few works consider the cross-receiver scenario. The performance deteriorates significantly when the model is deployed on new receivers directly. To this end, a cross-receiver RFF learning scheme is proposed. First, an unsupervised pre-training method based on contrastive learning is utilized to extract receiver-agnostic features. Then, the model is optimized by subdomain adaptation to further improve identification performance. The proposed scheme does not require multiple labeled datasets from different receivers. And experimental results indicate that the proposed scheme effectively alleviates performance degradation in the cross-receiver scenario.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据