基于相像系数与SVM的雷达辐射信号分选技术  

Sorting Radar Emitter Signal Based on Resemblance Coefficient and SVM

作　　者：刘江波[1] 

机构地区：[1]91404部队,秦皇岛066000

出　　处：《电子对抗》2014年第2期25-28,共4页Electronic Warfare

摘　　要：提出一种基于相像系数与二叉树支持向量机的雷达辐射源分选新方法。通过构造基本序列，提取辐射源信号与其的相像系数得到分选特征向量。采用聚类分析中的均值距离来生成二叉树，将分类器分布在各个节点上，构成了多分类支持向量机，减少了分类器数量和重复训练样本的数量。仿真实验结果表明，该方法能用较少的特征获得较高的正确识别率。In this paper, a new method of radar emitter signal recognition based on resemblance coefficient and the binary tree Support Vector Machine （SVM） is proposed. The resemblance coefficient of radar emitter signals is extracted and it is used as sorting characteristic. It uses average class distance of clustering to construct binary tree. The method distributes classifiers each node which constitutes multi-class SVM, so it can reduce the number of SVM classifiers and repetitive training samples. Simulation experiment results show that the introduced method achieves good accurate recognition rate in terms of a little features as reference.

关 键 词：相像系数 支持向量机 二叉树 雷达辐射源 

分 类 号：TN95[电子电信—信号与信息处理]