Airborne Multi-function Radar Air-to-air Working Pattern Recognition Based on Bayes Inference and SVM

Traditional identification methods often depend on the validity of training data set and the rationality of parameter selection, which leads to the decrease of availability. A comprehensive recognition method based on Bayes reasoning and SVM classifier is proposed in this paper to address the difficulty of radar operating pattern recognition under non-cooperative confrontation and jamming pulse conditions. According to the tactical application characteristics and hierarchical structure of radar operation mode, a feature parameter extraction method based on CPI is constructed. And the pattern recognition rate Bayes inference algorithm is improved base on the SVM algorithm. Simulation results show that the accuracy of this method is improved by 1.37% on average, and is 98.28% and 92.79% respectively under cooperative and non-cooperative confrontation, which proves the effectiveness of the algorithm.