Method and apparatus for active audio matrix decoding

A method and an apparatus for an active audio matrix decoding are provided to tune an optimum level of each channel signal based on a position of a virtual sound image generated based on a practical environment. A method for an active audio matrix decoding includes the steps of: decoding a stereo signal into a multi-channel signal(1010); extracting a power vector of a signal according to each channel by multiplying a speaker position of the plurality of channels to a size of each decoded channel signal(1020); extracting a vector of a virtual sound source between each channel by linearly coupling power vector values of each channel extracted at the above step(1030); normalizing positions of each channel speaker based on the position of the dominant sound image(1050); and distributing a gain value to each normalized channel by comparing a signal size of the decoded whole channel with the signal size of each channel(1060).