Abstract: In order to recognize the deceptive jamming mode to make radar more effective in chosing anti-jamming measures, a deceptive jamming recognition method based on bispectrum feature analysis and pattern recognition technology is proposed. In this method, the bispectrum distribution and estimation method of deceptive jamming is given firstly, based on its distribution feature and statistical characteristic, two feature factors of deceptive jamming are defined, and then the support vector machine (SVM) classifier based on kernel clustering algorithm is designed to classify the feature factors and to recognize the different deceptive jamming modes. Finally, the complete jamming mode recognition model is built and simulated. The results of simulation show that this model has high correct recognition rate for 3 different deceptive jamming modes and little affected by jamming to noise rate.