基于双域特征的雷达欺骗干扰样式识别方法

Method of Radar Deceptive Jamming Modes Recognition Based on Two Domains Features

数字储频技术和DSP处理芯片技术的发展大大提高了欺骗干扰技术的实时性和自适应性,对欺骗干扰信号样式的识别更加困难。针对这一问题,提出了一种新的雷达欺骗干扰样式的识别方法:梳理了几种欺骗干扰样式的数学模型,对信号作快速傅里叶变换后定义信号特征平稳度,在对来波信号进行双谱估计的基础上定义信号特征凸度,构建涉及频域、双谱域的"平稳度、凸度"二维特征空间,设计GA-BP神经网络分类器对信号具体欺骗干扰样式进行识别。仿真实验表明,该识别模型具有较好的正确识别率和实时性,且受干噪比影响小。

With the development of digital radio frequency memory （DRFM） technique and digital signal processor （DSP） processing chip technique,the real-time and self-adaption of the deception jamming mode has improved greatly and recognizing the deceptive jamming mode is becoming more and more difficult. In order to recognize the deceptive jamming mode,a new method of deceptive jamming recognition is proposed. In this method ,the mathematical model of three deceptive jamming modes is given firstly. Based on the fast Fourier transform and bispectrum estimation,two feature factors, including steady degree and protruding degree, are defined. Then feature space of 2 dimension covering frequency domain and bispectrum domain is constructed. Finally the genetic algorithm-back propagation （GA-BP） neural network classifier is designed to recognize the deceptive jamming modes. The results of simulation show that the recognizing mode performs well in the correct recognition rate and the aspect of real-time. In addition, the model is little affected by jamming to noise rate（JNR）.