Title :
Independent vector analysis, the Kotz distribution, and performance bounds
Author :
Anderson, Matthew ; Geng-Shen Fu ; Phlypo, Ronald ; Adali, Tulay
Author_Institution :
Univ. of Maryland Baltimore County, Baltimore, MD, USA
Abstract :
The recent extensions of independent component analysis (ICA) to exploit source dependence across multiple datasets, termed independent vector analysis (IVA), have thus far only considered two multivariate source distribution models: the Gaussian and a second-order uncorrelated Laplacian distribution. In this paper, we introduce the use of the Kotz distribution family as a more flexible source distribution model which exploits both second and higher-order statistics. The Cramér-Rao lower bound (CRLB) for IVA performance prediction is shown to be analogous to the bound for blind source separation (BSS). Lastly, we provide an analytic expression for the CRLB when the sources follow the multivariate power exponential (MPE) subclass of distributions within the Kotz family.
Keywords :
Gaussian distribution; blind source separation; higher order statistics; independent component analysis; BSS; CRLB; Cramer-Rao lower bound; Gaussian distribution model; ICA; IVA performance prediction; Kotz distribution; MPE distribution subclass; blind source separation; higher-order statistics; independent component analysis; independent vector analysis; multivariate power exponential subclass; multivariate source distribution model; second-order statistics; second-order uncorrelated Laplacian distribution model; source dependence; Algorithm design and analysis; Blind source separation; Brain models; Covariance matrices; Independent component analysis; Vectors; Cramér-Rao lower bound (CRLB); Independent vector analysis (IVA); Kotz distribution; joint blind source separation (JBSS); multivariate generalized Gaussian distribution; multivariate power exponential (MPE) distribution;
Conference_Titel :
Acoustics, Speech and Signal Processing (ICASSP), 2013 IEEE International Conference on
Conference_Location :
Vancouver, BC
DOI :
10.1109/ICASSP.2013.6638257