基于VMD和特征融合的辐射源信号识别

Radiation emitter signal recognition based on VMD and feature fusion

在日趋复杂的电子对抗中,如何提高雷达辐射源信号(radar emitter signal,RES)识别率和抗噪性能是亟待解决的问题。为此提出了一种变分模态分解(variational mode decomposition,VMD)和特征融合相结合的RES识别方法。首先利用VMD算法对各雷达信号进行分解得到3个本征模态函数(intrinsic mode function,IMF);然后,对这3个IMF分量提取排列熵(permutation entropy,PE)和样本熵(sample entropy,SE)特征进行特征融合,构成六维特征向量;最后利用支持向量机对辐射源信号进行识别。利用6种不同的辐射源信号对该方法进行了验证,仿真实验结果表明,该方法在低信噪比(signal to noise ratio,SNR)下能取得较好的识别率,当SNR不低于0 dB时,六维特征向量的识别率达到100%,具有较强的抗噪性能。

In the increasingly complex electronic warfare,how to improve the radar emitter signal(RES)recognition rate and anti-noise performance is an urgent problem to be solved.For this reason,a RES recognition method based on a combination of variational mode decomposition(VMD)and feature fusion is proposed.Firstly,the VMD algorithm is used to decompose the radar signals to obtain three intrinsic mode functions(IMF).Then the permutation entropy(PE)and sample entropy(SE)features of the three IMF components are extracted to form a 6-dimensional feature vector.The support vector machine is used to identify the radar emitter signals.The method is validated by six different radar emitter signals.The simulation results show that the proposed method can achieve the better recognition rate under low signal to noise ratio(SNR).When the SNR is not lower than 0 dB,the recognition rate of the 6-dimensional feature vector reaches 100%,and it has strong anti-noise performance.