多特征融合的雷达信号脉内调制识别

Fusion of characters for intra-pulse modulation recognition of radar signal

针对当前电子侦察情报系统自动识别率低、识别类型少的问题,提出了一种有效的联合奇异值分解和主分量分析的雷达信号脉内调制类型自动识别方法。该方法对信号时频图像进行奇异值分解(SVD)和主分量分析(PCA)后,将所得特征参量按一定的准则进行融合来识别信号。仿真结果表明,SVD和PCA相融合的识别方法在低信噪比(4 dB)下,对常见脉内调制信号正确识别率均大于90%,并且该方法分类器具有结构简单、抗噪声能力强的优点。

Aiming at the problem of low automatic recognition rate and less types in electronic reconnaissance system,an effective method for automatic recognition of intra-pulse modulation type based on fusion of Singular Value Decomposition（SVD） and Principal Component Analysis（PCA） was put forward.After SVD and PCA to Time Frequency Distribution（TFD） image of the radar signal,the method fused eigenvectors based on certain fusion rules to recognize radar signal.The results of simulation indicate that the method based on fusion of SVD and PCA is of higher correct recognition more rate than 90% under the low SNR（4 dB）.In the meantime,the structure of classifier is simple and not sensitive to noise.