The invention provides a radiation source fingerprint feature extraction method based on wavelet entropy and chaotic features, and relates to the radiation source individual identification field. Theradiation source fingerprint feature extraction method based on wavelet entropy and chaotic features includes the following steps: 1) performing extraction of wavelet entropy features on radiation source signals to obtain feature parameters; and 2) performing analysis of chaotic features on the radiation source signals, extracting the chaotic feature parameters, combining the wavelet entropy feature vector with the chaotic feature vector to obtain the combination feature vector, inputting a feature classifier to realize individual identification of the radiation source. The radiation source fingerprint feature extraction method based on wavelet entropy and chaotic features overcomes the defect that at present a commontime frequency analysis method does not consider the non-linear nature ofthe radiation source signals, gives play to the strong time frequency resolution characteristic of wavelet packet transformation, extracts the multi-scale local features of the signals, and considersthe non-linear situation of the signals by means of chaotic analysis which is a non-linear analysis method so as to more accurately reflect the features of the radiation source signals and enable theextracted feature parameters to be higher in discrimination performance.