TY - GEN
T1 - Convolutive blind source separation by fourth-order statistics
AU - Wang, Fuxiang
AU - Zhang, Jun
PY - 2009
Y1 - 2009
N2 - In this paper, we present a new method using fourth-order statistics for convolutive blind source separation. After a whitening process of the observed data, the separation of convolutive mixtures are transformed to find a semi-unitary demixing matrix. Then the semi-unitary demixing matrix is determined by joint diagonalization for a set of fourth-order joint cumulant matrices. Simulation results of speech separation demonstrate the effectiveness of the new approach.
AB - In this paper, we present a new method using fourth-order statistics for convolutive blind source separation. After a whitening process of the observed data, the separation of convolutive mixtures are transformed to find a semi-unitary demixing matrix. Then the semi-unitary demixing matrix is determined by joint diagonalization for a set of fourth-order joint cumulant matrices. Simulation results of speech separation demonstrate the effectiveness of the new approach.
KW - Blind source separation
KW - Fourth-order statistics
KW - Joint diagonalization
UR - http://www.scopus.com/inward/record.url?scp=77949587842&partnerID=8YFLogxK
U2 - 10.1109/ICICS.2009.5397743
DO - 10.1109/ICICS.2009.5397743
M3 - Conference contribution
AN - SCOPUS:77949587842
SN - 9781424446575
T3 - ICICS 2009 - Conference Proceedings of the 7th International Conference on Information, Communications and Signal Processing
BT - ICICS 2009 - Conference Proceedings of the 7th International Conference on Information, Communications and Signal Processing
T2 - 7th International Conference on Information, Communications and Signal Processing, ICICS 2009
Y2 - 8 December 2009 through 10 December 2009
ER -