雷达辐射源信号聚类分选算法综述

A Survey of Clustering and Sorting Algorithms for Radar Source Signals

现代战场环境下雷达信号密集、交叠严重,传统的雷达信号分选算法已不能有效完成未知雷达信号分选工作,而聚类分选因具有无监督学习、对先验知识要求小等优点被学者广泛应用于未知雷达信号聚类分选的研究工作。作者先总结了国内学者在传统聚类算法及其优化算法对未知雷达辐射源信号进行分类分选的部分理论研究成果,分析了各个算法存在的利弊,再根据雷达信号的特征参数以及影响聚类分选的外界因素提出了选择最优聚类算法的参考标准和建议。

In modern battlefield environment,radar signals are often dense and overlapped.The traditional radar signal sorting algorithms have not been able to complete the unknown radar signal sorting work effectively.Cluster sorting is widely used in the research of cluster sorting unknown radar signals because of its unsupervised learning and small requirements for prior knowledge.The author first summarizes some theoretical research results of domestic scholars in the classification and sorting of unknown radar emitter signals by traditional clustering algorithm and its optimization algorithm and analyzes the advantages and disadvantages of each algorithm,and then puts forward the reference criteria and suggestions for selecting the optimal clustering algorithm according to the characteristic parameters of radar signals and cluster influencing factors.