Research On Radar Signal Sorting Technology Based On Intra-pulse Features

The research and development of radar emitter signal sorting technology is of great significance for modern electronic warfare.With the development of radar technology,new system radar has been put into use extensively.As a result,electromagnetic environment becomes more and more complex,radar signal parameters are variable and the interception rate is low,which makes the traditional signal sorting methods difficult to obtain the desired effect.Based on the characteristics of radar signal intra-pulse modulation,this paper deeply studies the methods of signal pulse extraction,and carries out simulation analysis and verification,the main contents are as follows:(1)According to the ambiguity function of radar signal,the two-dimensional characteristic of ambiguity function is obtained by projection of time delay plane.A feature extraction method based on the symmetric Holder coefficients is proposed from the view of shape of the feature map.The obtained characteristic parameters are used to represent the intra-pulse feature of radar signal.A feature extraction method based on K-L transform is proposed from the view of data of the feature map,and the extraction of the signal intra-pulse feature by using the two-dimensional feature of the ambiguity function to construct the sample space.(2)The classical K-means clustering and sorting algorithm is studied,and the problems of the algorithm are analyzed in detail according to the algorithm flow.For the problem of selecting the initial cluster center,this paper deeply studies the K-means clustering algorithm based on density function,and analyzes the lack of algorithm.Then,the calculation method of density function is proposed,which improves the accuracy of clustering algorithm.In this paper,the ambiguity function analysis and the intra-pulse feature extraction algorithm are validated and analyzed by the simulation experiment to prove the feasibility of the method.The feature extraction method is evaluated by comparing the distribution of several characteristic parameters at different SNR.The optimal clustering algorithm is used to realize the signal sorting based on the intra-pulse feature,and the accuracy of the signal sorting is analyzed to prove the validity of the extracted characteristic parameters.