基于时频图像处理提取瞬时频率的雷达信号识别  

Instantaneous Frequency Based Radar Signals Recognition Using Time Frequency Image Processing

作　　者：朱健东[1,2] 张玉灵[3] 赵拥军[1] 

机构地区：[1]解放军信息工程大学 [2]63880部队 [3]郑州升达经贸管理学院

出　　处：《系统仿真学报》2014年第4期864-868,共5页Journal of System Simulation

基　　金：国家863项目(2011AA7031015);河南省科技攻关项目(122102210387)

摘　　要：时频图像可以刻画雷达信号的脉内调制方式,但是直接利用时频图像识别信号存在特征维数高,受噪声影响大等缺点。针对雷达信号识别问题,提出了一种利用Choi-Williams时频图像处理提取信号瞬时频率特征的识别方法。该方法从信号的时频分布出发,将信号的时频分布看成灰度图像,利用一系列形态学图像处理方法对图像进行降噪,提取时频图像中行索引信息作为信号的瞬时频率特征用于信号识别。仿真结果表明,通过图像处理方法提取的特征在低信噪比下具有很好的稳定性,在信噪比大于0dB时对7种典型的雷达信号识别率均超过90%,与现有方法相比,在低信噪比下就有更好的识别效果。Time-frequency image can represent intra-pulse modulation mode of radar signals. But the direct utilization of time frequency image to recognizing suffers from the high character dimension and noise. To classify radar signals, a novel recognition method was proposed, which used image processing to extract instantaneous frequency character from Choi-Williams time-frequency image. The approach regarded the time-frequency as a gray image. After a series of morphologic image processing, the row index information of the time-frequency image was extracted to represent the instantaneous frequency of radar signals with different modulation. The simulation results show the extracted character is very stable in low SNR, and the recognition rate of seven kinds of typical radar signals is more than 90% even when SNR ≥0dB. Compared with the existing methods, the proposed method has better recognition effect in low SNR.

关 键 词：雷达信号识别 时频图像 形态学处理 瞬时频率 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]