基于频响特性的大起伏密集假目标干扰识别技术

Recognition of Dense False Target Jamming with Large Fluctuations Using Frequency Response Characteristics

转发式密集假目标干扰具有压制性和欺骗性干扰的特点,干扰信号与目标回波极为相似,识别难度大。基于密集假目标干扰产生机理和射频链路物理特性,系统研究并提出雷达和干扰机频率响应特性及其对真、假目标回波幅-频映射特征的影响机制模型;针对干扰信号功率动态范围较大的实际情况,提出大动态信/干噪比条件下基于频响特性的密集假目标干扰识别方法;通过卷积神经网络-长短时记忆网络双通道特征融合的基分类器构建和M/N逻辑分类器集成,实现真-假回波信号频响起伏特征提取与分类识别。研究结果表明:所提方法在实测数据上获得了94.5%以上的识别精度,证明了其有效性和先进性;所做工作对大起伏密集假目标干扰识别具有参考意义。

Forwarding dense false target jamming is suppressive and deceptive jamming,with jamming signals quite similar to those of the target echo,thus posing challenges for recognition.Based on the generation mechanism of dense false target jamming and the physical characteristics of the Ratio Frequency link,the frequency response characteristics of radar and jammer and the mechanism model of their influence on the amplitude-frequency mapping characteristics of true and false target echoes are systematically studied and proposed.On this basis,leveraging the large dynamic range of jamming signal power,a method of dense false target jamming recognition based on frequency response characteristics and large dynamic SNR/JNR is proposed.Through the construction of basis classifier with a dual-channel feature fusion network comprising convolutional neural network and long short-memory network(ODCNN-LSTM),the M/N logical criteria is used to integrate the basis classifiers.Then,feature extraction and recognition of true and false echo signal frequency response fluctuations are realized.The recognition accuracy is over 94.5%for the measured data,demonstrating the effectiveness and innovation of the proposed method.This work holds significance for recognizing dense false target jamming with significant fluctuations.