Research On Sorting Method For Unknown Radar Emitter Signals

With the development of science and technology, radars with new system and technology appear continually. Because of the increasing complex in military radar’s signal form, high in density, large in duty ratio and variable in waveform parameters, and the influence of the electromagnetic environment on a battlefield, the radiation source signal received by a radar reconnaissance system has high density and huge number of pulses. The traditional classification method based on five parameters and its improved methods have encountered serious difficulties in sorting new radar emitter signal. Therefore, it is a key and difficult research topic of modern electronic countermeasures for exploring an effective radar emitter sorting method which can sort all kinds of radar emitter signals in a complex battlefield electromagnetic environment.In this thesis, a comprehensive sorting method for unknown radar emitter signal under complex electromagnetic environment is studied. Firstly, the basic theory of radar emitter signal sorting and recognition is briefly introduced. According to the parameter characteristics of radar emitter signal in radar reconnaissance system, the radar emitter signals and their characteristics are classified. Besides, a detailed summary of the advantages and disadvantages of radar emitter signal sorting is made. The key factors affecting radar emitter signals’ separation are analyzed.Based on DOA feature parameters as pulse dilution, the K-Means clustering algorithm and the FCM clustering algorithm are compared to apply the FCM clustering algorithm to the whole separation process. Besides, combinition the polarization characteristic with RF and PW is proposed in the radar emitter signal’ separation.Modern radars can promptly change its parameters in the pulse and it has various work mode. It is necessary to search for new and stable clustering parameters. Intra pulse modulation feature parameters are relatively stable, thus it is feasible to separate radar radiation source signal based on single pulse. Moreover, they are insensitive to noise, and perform well in aggregation degree within class and separation between classes. Through simulated analysis, the features of Cr2、Wpt4 and Wpt7 in the pulse modulation are used as the main sorting parameters.Based on the work above, this thesis presents a synthetic classification scheme of radar emitter signals under complex electromagnetic environment. And the whole process is simulated by MATLAB. The experimental results show that the proposed method can effectively accomplish the accurate separation of multiple radar emitter signals in complex electromagnetic environment.