摘要
针对当前通信辐射源个体识别方法存在识别效率不高,内在细微特征难以提取等问题,提出了基于双谱特征融合的通信辐射源特征提取算法。该算法通过融合对角积分双谱与双谱对角切片特征组成特征向量,并对对角切片特征提取方法进行了改进,将原信号先进行经验模态分解(EMD),在得到的本征模函数(IMF)基础上提取对角切片特征,最后使用支持向量机(SVM)得到分类结果。分类实验结果表明,基于双谱特征融合的算法较积分双谱的算法在识别效果上有一定程度的提升。
In view of the problem that the identification method is not high and the intrinsic feature is difficult to be extracted,a feature extraction algorithm based on the bispectrum feature fusion was proposed .The feature vector was formed by the combination of diagonal integral bispectrum and the feature of bispectrum slice,and the original signal was decomposed by empirical mode decomposition;then the feature of the bispectrum slice was extracted from the instrinsic mode functions;finally the support vector machine was used to obtain the clas-sification results.Six station classification experimental results showed that the proposed algorithm based on the bispectrum feature fusion integration was better.
出处
《探测与控制学报》
CSCD
北大核心
2016年第5期91-95,共5页
Journal of Detection & Control
关键词
通信辐射源
特征提取
对角积分双谱
双谱切片
communication transmitter
feature extraction
diagonal integral double spectrum
bispectrum slice
作者简介
桂云川（1991-）,男,江西鹰潭人,硕士研究生,研究方向：通信辐射源特征提取.E-mail：15209837812@163.com.
