Post date: Oct 05, 2014 1:6:23 PM
The problem of separating sources without a priori knowledge of their number or spatial distribution is known as ”blind separation”.Most existing techniques for approaching the problem of source separation employ Independent Component Analysis (ICA) algorithm with information maximization We used an MDCT approach , signal representation in the modified discrete cosine transform(MDCT) domain has emerged as a dominant tool in high quality audio coding because of its special properties.In addition to an energy compaction capability, MDCT simultaneously achieves critical sampling and MDCT is a real transform which is based on the type-IV discrete cosine transform (DCT-IV). Namely, the MDCT coefficients are real and its phase can be considered as a binary value. The phase will not be changed unless the additive noise changes the sign of MDCT coefficients. Thus it can minimize the distortion of the erroneous phase