Intelligent Recognition Methods For Radar Emitter Signals

In the electronic warfare environment, the radar emitter signal recognition is not only the purpose of electronic measures, but also the basis to judge the threat of enemy weapons. It is one of the key procedure of signal processing in Electronic Intelligence (ELINT), Electronic Support Measures (ESM) and Radar Warning Receiver (RWR) systems, and plays an important role in the entire electronic warfare.Feature extraction and intelligent recognition for advanced radar emitter signals are studied exploringly in this thesis. In the first part, feature extraction is given. The application of time-frequency analysis and wavelet analysis in gaining the features of intra-pulse modulation are introduced. The improved algorithm which obtained the feature vectors of the instantaneous frequency is proposed, and some relative simulations are done to the relevant technologies. In the second part, intelligent recognition is researched. The algorithms of unsupervised clustering and supervised classification are introduced. According to the extracted feature vector, we identify radar emitters under the unsupervised clustering and supervised SVM. The results of the simulation indicate that the SVM classification under the conditions of small samples can still achieve good results.