Recognition Method of Radar Signal Based on Rough Set and Support Vector Machine

A hybrid algorithm based on attributes reduction of rough set and classification principles of support vector machine (SVM) is presented in this paper. Firstly, the attributes reduction of rough set has been applied as preprocessor so that we can delete the redundant attributes and conflicting objects from decision making table but remain efficient information lossless. Then, the classification modeling and forecasting test based on SVM are realized. By this method, the dimension of data is reduced greatly, the complexity in the process of SVM classification is decreased highly, the occupied memory is cut down and the over-fit of training model is prevented at some extent, also the good classification performance is obtained. Finally, the simulation experiment of radar signal recognition and its results show this hybrid method is effective.