Abstract: In order to solve the problem of individual identification of communication radiation sources under non-cooperative communication conditions, a specific emitter identification method using wavelet transform feature extraction is proposed. The method performs wavelet transformation on the received non-cooperative communication signal, and select the optimal wavelet basis which is used to extract feature vectors by calculating the degree of separation between classes, then according to the feature distribution select the wavelet coefficient complexity under specific wavelet basis as the SNR reference value to assist individual identification. The simulation results show that the recognition effect is better in the environment where the SNR changes and the individual differences of communication radiation sources are small, thus verifying the effectiveness of the method.