Concerning the modulation recognition problem of the single-channel receiver working under the multi-signal environment,a new approach which can directly extract the feature of each independent component was proposed to identify multi-component Phase Shift Keying(PSK) signals.In this method,the spectral correlation function of the multi-component signal was deduced,features of signals were extracted and the classifier was designed.The theoretic analysis and experimental results show that the proposed algorithm can extract the feature of each independent component under the mutual interference of component signals,it can effectively classify arbitrary combinations of dual-signals in the given set,and the average correct recognition rate of all combinations reaches 97% when Signal-to-Noise Rate(SNR) is 0 dB.