基于HOS的未知雷达辐射源信号分选  

Signal Sorting for Unknown Radar Emitter Based on HOS

作　　者：杨志祥[1] 朱元清[1] 徐光华[2] 任静波 

机构地区：[1]空军雷达学院电子对抗系,武汉430019 [2]驻中国电子科技集团公司38所军代室,合肥230031 [3]解放军93993部队,兰州7730102

出　　处：《电子对抗》2008年第6期10-13,共4页Electronic Warfare

摘　　要：对未知雷达辐射源信号进行准确分选是电子对抗领域迫需解决的一个难题。当前用于分选的特征参数主要有相像系数、小波包变换、熵值等，它们普遍存在的问题是在低信噪比下难以获得令人满意的准确率。基于高阶统计量（HOS）可以实现多种不同雷达辐射源信号的分类，且具有抑制高斯噪声的优点。文章提出一种基于HOS的未知雷达辐射源信号分选新方法，通过计算机仿真验证，在相同的信噪比条件下，利用HOS进行分选可以获得更高的准确率。It is a difficult problem to sorting unknown radar emitter in the electronic countermeasure field. At present, the characteristic parameters which are used to signal sorting are resemblance coefficient, wavelet package transformation and entropy, but all of them can＇ t get a high right rate at low SNR. Many different kinds of modulation signals can be sorted by high order statistic（HOS）, and also the gauss noise can be restrained. A new method based on HOS is proposed in this paper, simulation results show that the new method can get a higher right rate than any other methods in the same SNR condition.

关 键 词：高阶统计量(HOS) KOHONEN神经网络 分选 

分 类 号：TN95[电子电信—信号与信息处理] TN971.1[电子电信—信息与通信工程]