支持向量机在雷达辐射源信号识别中的应用  

Application of Support Vector Machine to Radar Emitter Signal Recognition

作　　者：张葛祥[1] 荣海娜[1] 金炜东[1] 

机构地区：[1]西南交通大学电气工程学院,四川成都610031

出　　处：《西南交通大学学报》2006年第1期25-30,共6页Journal of Southwest Jiaotong University

基　　金：国家自然科学基金资助项目(60572143);国防科技重点实验室基金资助项目(NEWL51435QT220401)

摘　　要：为了提高电子对抗设备的信号识别能力，采用相像系数法提取雷达辐射源信号特征，并引入支持向量机完成信号自动分类识别．相像系数法在大信噪比范围内稳定性好、分辨能力强．支持向量机分类器结构简单、可获得全局最优、泛化能力强．实验结果表明，基于相像系数和支持向量机的辐射源信号识别方法在大信噪比（5～20dB）范围内，错误识别率最低可达2．68％，优于传统识别方法．To enhance the ability of electronic warfare equipment to recognize signals, resemblance coefficient method was proposed to extract features from radar emitter signals, and support vector machine （SVM） was introduced to identify different signals automatically. Resemblance coefficient features have good stability and discriminability. SVM has good characteristics of simple structure, global optimum and strong generalization ability. Experimental results show that the introduced approach for recognizing radar emitter signals using resemblance coefficient and SVM is superior to the conventional ones. It works effectively in a large range of noise to signal ratio （5 to 20 dB） with the recognition error rate being as low as 2.68%.

关 键 词：模式识别 信号处理 支持向量机 相像系数 雷达辐射源 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程] TN957[自动化与计算机技术—控制科学与工程]