Radar Signal Sorting Method Based On PRF And Intra-pulse Modulation

Radar signal sorting is an important part of electronic intelligence reconnaissance.In today's increasingly complex electromagnetic environment,various kinds of complex reconnaissance environment and engineering requirements emerge in endlessly.The existing sorting and identification technology is facing new problems and challenges.In this paper,the problem of radar signal sorting in complex electromagnetic environment is discussed.The main lobe pulse correlation sorting,clustering of emitters,data analysis and software implementation are studied respectively.The main work of this paper is as follows:(1)In the pulse radar signal sorting,many identical targets only intercept radar main lobe pulse,which brings great difficulty to accurate sorting.A method of radar main lobe pulse correlation sorting is proposed,which uses pulse repetition frequency to extract main lobe pulse and correlative clustering to realize signal sorting.To achieve better separation of the same pulse repetition interval(PRI)emitter signal in the main lobe detection environment with high pulse loss rate.(2)This paper studies the problem of signal clustering based on the characteristics of intra pulse unintentional modulation.Since the limitations of the existing algorithm and the characteristics of the radar signal in pulse unintentional modulation,an improved density peak clustering algorithm based on kernel density estimation is proposed.First,the optimal bandwidth is estimated by kernel density to select the adaptive truncation distance,then the difference between the local density and the relative distance of each point is calculated,and then the critical point is found by the change of slope to determine the clustering center.The simulation and experimental data show the algorithm can effectively improve the shortcomings of the traditional density peaks clustering algorithm(DPC),and it can be better applied in the clustering and sorting of the unintentional modulation characteristics of radar signals.(3)Clustering experiments of radar emitter characteristic parameters based on measured signals are carried out.The clustering performance of different feature parameters and different clustering algorithms is examined,and the effectiveness of the improved DPC clustering algorithm proposed in this paper is verified.The possibility of radar emitter feature parameters clustering in engineering application is explored.(4)Aiming at the engineering requirement of radar emitter individual classification,the software of radar emitter individual classification is designed and implemented.The automatic processing of radar emitter unintentional modulation features is realized in new target learning,new target recognition and existing target recognition.The validity of the improved DPC clustering algorithm proposed in this paper is verified in engineering practice.