基于双谱分析的雷达有源欺骗干扰识别

Active Deception Jamming Identification Method Based on Bispectrum Analysis

针对跟踪雷达的三种常见欺骗干扰识别算法计算量大,使用范围较窄等问题,提出了基于双谱分析的识别算法。该算法对受到干扰与否的四种情况下的雷达接收信号进行双谱分析,经过降维和归一化处理后,将三维双谱信息转化为二维特征信息,对得到的二维特征谱提取熵特征和盒维数组成特征参数集放入分类器中进行分类识别。仿真实验表明,在SNR=4dB时,三种干扰的识别率均大于90%,在SNR=6dB时,三种干扰的识别率均接近100%。证明了该方法有较高的识别率,特别是该方法受信噪比影响较少,能够有效降低噪声对干扰识别的影响。通过与其他文献方法进行比较,证明了该方法的优越性。

In order to enhance the tracking radar operational effectiveness,a recognition algorithm was proposed based on bispectrum analysis in allusion to heavy calculation and narrow range of the major three types of deception jamming identification.In this method,bispectrum analysis of received radar signal under four kinds of jamming was given firstly,and after a series of dimensionality reduction and normalization,a three-dimensional feature information was turned into two-dimensional feature information.which were calculated as the feature vector for signal recognition.Box-counting and entropy were calculated as the feature of a two-dimension feature spectrum which constituted feature vector for signal recognition.Experimental results showed that the proposed approach could achieve satisfying recognition and its identification rate was over 90%when the signal-to-noise ratio is 4 dB,its identification rate was over 100%when the signal-to-noise ratio was 6 dB,Particularly the method was less affected by SNR.