Research On Jamming Technology For Radar Target Recognition Of One-dimensional Range Image

With the increasing of radar signal bandwidth,radar target recognition technology develops rapidly.At present,broadband radar mainly classifies detection targets through onedimensional range image,Synthetic aperture radar(SAR)image and inverse synthetic aperture radar(ISAR)image.with the rapid development of one-dimensional range image target recognition technology,The research of jamming technology for target recognition of radar one-dimensional range image has become a new research direction.In this thesis,the jamming technology of radar one-dimensional range image target recognition is studied.The jamming performance of three kinds of covering jamming to onedimensional range image target recognition algorithm is compared and analyzed,and false target jamming based on time delay and false target jamming based on smart noise are studied,and a smart noise jamming based on comb spectrum which can generate multiple false targets is proposed,and verified by simulation experiment.Firstly,the technology of radar target recognition in one dimension range image is studied.This paper mainly tease out the imaging principle and sensitive characteristics of onedimensional range image,explains the feature extraction method of one-dimensional range image,and the principle of Mahalanobis distance classifier,Bayes classifier,Support Vector Machines(SVM)classifier and Convolutional Neural Network(CNN)classifier.Three kinds of target one-dimensional range image are simulated by FEKO.And the recognition of three kinds of targets by these four kinds of classifier is calculated.Secondly,the covering jamming performance of the four kinds of classifiers is studied.Covering jamming for one dimensional distance image recognition is studied.The mechanism of covering jamming and the mathematical model of jamming signal are discussed.Through simulation experiments,covering jamming signals with different jamto-signal ratios are studied,and the variation of the recognition rate of the four kinds of classifiers for the three kinds of targets is calculated and the comparative analysis is made.Finally,the deception jamming for radar one-dimensional range image target recognition is studied,put forward for false target jamming of based on time delay of one dimensional distance for classifier,on the theoretical derivation and simulation,the experimental results show that this kind of interference pattern parameter Settings is complex,and the classifier of deception jamming effect is not ideal;Therefore focuses on one-dimensional distance as the process of feature extraction in target recognition,is proposed based on smart noise of one-dimensional distance as false target jamming,on the theoretical derivation,and illustrates the implementation process,and the feasibility of clear interference is validated by computer simulation,and then the false target based on smart noise jamming is improved,A smart noise jamming based on comb spectrum is proposed,which can generate multiple false targets.The feasibility of jamming is proved by theoretical derivation,and the realization process is illustrated.And the jamming is verified by simulation experiment.Based on the above studies,it is concluded that: For covering interference,noise frequency modulation jamming has the best performance for the four kinds of classifier under different jam-to-signal ratios.The false target jamming based on smart noise can generate a single pseudo target one-dimensional range image,one dimensional range images of multiple false targets are generated by smart noise jamming based on comb spectrum.It has certain engineering reference value.