In order to improve the classification accuracy of radiation source signal in complex electromagnetic environment,a joint decision classification algorithm based on affinity propagation(AP)clustering and support vector machine(SVM)is proposed.The algorithm removes the eletromagnetic interference signal and get the in-itial number of categories and clusters of the spatial signal.To get better classification effect,the probability model-based SVM is employed to reclassify samples in clustering clusters.Then,the probability of the sample category and the sample numbers in the proportion of the original clustering cluster are combined to verify the category of the input signals and make a confirmation of the samples threat levels.Simulation results show that the proposed algorithm can improve the signal classification accuracy in the background of electromagnetic sig-nal characteristics overlap and radiation source of complex system effectively,and it can enhance the adaptabil-ity of weapon system in complex environment.
维普
万方数据