Single-channel mix signal separation technique in noise

The paper presented a new technique of signal processing to accomplish blind source separation when given a single-channel mixture signal in noise. One signal source can be generated by a set of weighted linear superposition of the time domain sub-band functions with independent component characteristic. By combining the independent sub-band function components into the single mixture signal, a single mixture signal can be transformed into a multi-dimensional vector from one-dimensional. And the separation model of the single-channel signal in noise is studied. We consider using wavelet transform to eliminate the leaving noise. The separation results of TEOAEs mixture signal demonstrated the effectiveness and adaptability of the proposed method.