摘要
面对复杂多变的战场电磁环境与日益革新的雷达技术,传统雷达辐射源识别技术由于灵活性差、对先验知识依赖严重等问题受到严峻挑战,基于机器学习的雷达辐射源识别技术具有更强的泛化性和智能性受到研究学者的广泛关注。首先针对基于机器学习的雷达辐射源识别技术的产生过程进行梳理,然后从统计学习、神经网络、迁移学习、集成学习、聚类5个方面综述了相关研究成果并对各自方法的性能进行了分析比较,最后针对该研究方向上亟待解决的问题和难点做了相关的探讨。研究成果将为基于机器学习的雷达辐射源识别技术在实际应用过程中起到参考和借鉴作用。
Faced with the complex and ever-changing battlefield electromagnetic environment and increasingly innovative radar technology, traditional radar emitter identification technology is severely challenged due to poor flexibility and serious dependence on prior knowledge. Radar emitter identification technology based on machine learning has stronger generalization and intelligence, which is widely concerned by researchers. Firstly, it combs the generation process of radar emitter identification technology based on machine learning, and then summarizes the related research results from five aspects: statistical learning, neural network, transfer learning, ensemble learning and clustering. The performance of each method is analyzed and compared. Finally, the related problems and difficulties to be solved in this research direction are discussed. The analysis of the paper will provide reference and advice in the practical application of radar emitter identification technology based on machine learning.
作者
李昆
朱卫纲
Li Kun;Zhu Weigang(Graduate School,Space Engineering University,Beijing 101416,China;Department of Electronic and Optical Engineering,Space Engineering University,Beijing 101416,China)
出处
《电子测量技术》
2019年第18期69-75,共7页
Electronic Measurement Technology
基金
CEMEE国家重点实验室项目(2018Z0202B)资助
关键词
机器学习
辐射源识别
智能化
machine learning
source of radiation identification
intellectualization
