针对低信噪比下雷达辐射源信号的分类识别,提出了一种将时频分析与图像处理相结合的特征提取和识别方法.该方法首先对雷达信号进行时频变换,将得到的时频分布转化为灰度图像；然后运用图像处理方法对时频图像进行增强和去噪；最后提取局部二值模式纹理特征描述子作为信号的识别特征,并采用支持向量机分类器实现信号的分类识别.文中针对12种常见雷达信号进行了仿真,结果表明该方法在较低的信噪比下仍能获得较为满意的识别率,当SNR=0dB时,信号的平均识别率能达到95.35％.所提出方法能有效降低噪声对分类识别的影响,同时对于时频图像相近的信号也有较好的识别效果,表明了该方法的有效性.%To correctly classify advanced radar emitter signals in the condition of low signal-to-noise ratio, a novel approach is proposed by using image feature of time-frequency distribution for radar emitter signal recognition, which transforms the classification of emitter signals into image processing and image recognition. Time-frequency images of radar emitter signals are obtained by using modified B distribution, and then these images are transformed into grayscale images. In addition, the time-frequency images are enhanced and denoised by image processing methods. Finally, the texture features of time-frequency image are extracted for signal recognition based on the local binary patterns, and the support vector machine is applied to identify radar emitter signals automatically. Simulation results show that the proposed approach can achieve satisfactory accurate recognition in low signal-to-noise rate (SNR). Even for SNR =0dB, the overall correct classification rate of twelve typical radar emitter signals is 95. 35%. The proposed approach can reduce the impact of noise effectively, and also distinguish the approximate time-frequency images well. The validity of the approach is demonstrated by experimental results.
展开▼
