基于SSA的DRFM速度欺骗干扰识别

DRFM Velocity Deception Jamming Recognition Based on Singular Spectrum Analysis

针对数字射频存储(DRFM)干扰机的转发式欺骗干扰的高度相干而难以识别的问题,本文提出一种基于奇异谱分析(SSA)的DRFM速度欺骗干扰识别方法。由于DRFM速度欺骗干扰信号因干扰机数控移相器的步进阶跃调相会产生谐波分量寄生,导致其SSA分解后奇异值分布与目标回波存在较明显的差异。因此本文首先利用SSA算法获取待识别信号的奇异值,然后提取奇异值的统计直方图的方差、峰度、偏度、能量和熵作为特征构建特征向量,最后利用支持向量机(SVM)作为分类器实现雷达目标与DRFM速度欺骗干扰信号的识别。该方法不再依赖现有的A/D相位量化的谐波模型,且在较低信噪比下能有效地识别出干扰信号。计算机仿真验证了该方法的有效性。

The deception jamming transmitted by a digital radio frequency memory(DRFM)jammer owns the characteristic of high coherence and is difficult to be recognized.To settle this problem,a novel method of recognizing the velocity deception DRFM jamming is proposed based on singular spectrum analysis(SSA).Due to the harmonic components in velocity deception jamming signals caused by the stepping phase modulation of digitally controlled phase shifter of DRFM jammer,the singular value energy distribution of the jamming signal has great difference from that of the target echo.Therefore,SSA is firstly used to obtain the singular values of the signal to be recognized.Then,the variance,kurtosis,skewness,energy and entropy of the histogram of the singular values are extracted to construct the feature vector.Finally,the classifier which is constructed by using support vector machine(SVM)is used to accomplish the recognition of DRFM-based velocity deception jamming.The proposed method does not depend on the existing harmonic model of analogue-to-digital(A/D)converter phase quantization,and can effectively identify the jamming at low signal-to-noise ratio.The validity of the proposed method is evaluated using experimental data.