Title :
A least-squares approach to joint-diagonalization of tensor with application to source separation
Author :
Moreau, Eric ; LUIGI, CHRISTOPHE DE
Author_Institution :
ISITV, Toulon Univ., La Valette Du Var, France
Abstract :
In this paper, we consider the problem of the joint-diagonalization of square tensors through a least-squares approach. This allows us to show that recent joint-diagonalization criteria used in source separation can be seen based on a least-squares criterion, thus showing their optimality. A gradient algorithm is also proposed and a computer simulation illustrate the algorithm.
Keywords :
gradient methods; least squares approximations; source separation; tensors; gradient algorithm; joint-diagonalization criteria; least-squares approach; source separation; square tensors; Computer simulation; Cost function; Independent component analysis; Matrix decomposition; Reactive power; Signal processing algorithms; Source separation; Statistics; Tensile stress;
Conference_Titel :
Signal Processing and Its Applications, 2003. Proceedings. Seventh International Symposium on
Print_ISBN :
0-7803-7946-2
DOI :
10.1109/ISSPA.2003.1224825
