电讯技术2024，Vol.64Issue(7)：1163-1174,12.DOI:10.20079/j.issn.1001-893x.231030001
基于深度学习的跨域辐射源个体识别综述
Cross-domain Specific Emitter Identification Based on Deep Learning:a Comprehensive Survey
李奇真 1刘佳旭 2梁先明 1龙慧敏 1董海 1曹广平 1李建清2
作者信息
- 1. 中国西南电子技术研究所,成都 610036
- 2. 电子科技大学 电子科学与工程学院,成都 611731
- 折叠
摘要
Abstract
Specific emitter identification based on deep learning has become one of the main research methods to identify and authenticate wireless devices.However,the specific emitter identification algorithms based on traditional deep learning cannot be directly applied to cross-domain(cross-channel conditions,cross-receivers,cross-receiving time,etc.)emitter identification scenarios,because the model trained with data from one domain will generally be less effective when reasoning on data from another domain.The existing methods of cross-domain specific emitter identification based on deep learning including contrastive learning,transfer learning,and domain adaptation are investigated.The open-source data sets related to cross-domain specific emitter identification are also organized and summarized.The problems and challenges of cross-domain specific emitter identification are analyzed,and the development trend and future research directions of cross-domain specific emitter identification are prospected,in order to provide assistance for the practical application of deep learning in specific emitter identification in complex electromagnetic environments.关键词
跨域辐射源个体识别/深度学习/域适应/开源跨域辐射源数据集Key words
cross-domain specific emitter identification/deep learning/domain adaptation/open-source cross-domain specific emitter dataset分类
信息技术与安全科学引用本文复制引用
李奇真,刘佳旭,梁先明,龙慧敏,董海,曹广平,李建清..基于深度学习的跨域辐射源个体识别综述[J].电讯技术,2024,64(7):1163-1174,12.
