An approach for intra-pulse feature extraction of radar emitter signals was proposed based on the multi-resolution characteristics of wavelet transform. It was efficient to obtain quantitative information from signals. The energy entropy from approximation coefficients of wavelet transform and the other energy entropy from inter-scale correlations denoise of detail coefficients were adopted together as two-dimensional feature vectors. Experiment results demonstrated that the features of ten typical radar emitter signals extracted by wavelet transform showed good performance of noise-resistance and clustering when Signal-Noise Ratio(SNR) was low.