摘要
提出了一种基于经验模态分解(EMD,empirical mode decomposition)的通信辐射源个体识别方法。首先采用EMD方法将稳态信号的主要信号成分与杂散成分分离开来;然后提取杂散成分的频域特征作为信号的细微特征;最后运用支持向量机(SVM,support vector machine)分类器对多个通信辐射源个体进行分类识别。无线网卡个体识别的实验结果表明,对同型号的辐射源个体,该方法可以取得较好的识别效果;尤其在较低信噪比下,与其他方法相比,该方法获得更高的识别率。
A method using empirical mode decomposition（EMD） for identifying transmitter individuals is proposed.First of all,empirical mode decomposition is adopted to separate the steady state signal into main signal component and stray components.Then,the spectrum feature of stray components is extracted,which is regarded as fine feature of the signal.Finally,support vector machine（SVM） is employed to identify transmitter individuals.The experiment results of WLAN Cards identification manifest that the method is effective for identifying the transmitters with the same model and that this method performances better（especially in low SNR） than others.
出处
《中国电子科学研究院学报》
2013年第4期393-397,417,共6页
Journal of China Academy of Electronics and Information Technology
关键词
EMD方法
杂散成分
支持向量机
辐射源个体识别
empirical mode decomposition
stray components
support vector machine
transmitter individual identification
