Unified parametric ICA algorithm for hybrid sources and its stability analysis

Fasong Wang; Hongwei Li; Rui Li; Lihua Fu

doi:10.1117/12.657660

4 January 2006 Unified parametric ICA algorithm for hybrid sources and its stability analysis

Fasong Wang, Hongwei Li, Rui Li, Lihua Fu

Proceedings Volume 5985, International Conference on Space Information Technology; 59852E (2006) https://doi.org/10.1117/12.657660
Event: International Conference on Space information Technology, 2005, Wuhan, China

Abstract

Independent component analysis (ICA) refers to extract independent signals from their linear mixtures without assuming prior knowledge of their mixing coefficients. The purpose of this paper is to develop a novel unified parametric ICA algorithm, which enable to separate hybrid source signals including symmetric and asymmetric sources with a self-adaptive score functions. It is derived from the parameterized asymmetric generalized Gaussian density (AGGD) model. The parameters of the score function in the algorithm can be chosen adaptively by estimating the high order statistics of the observed signals online. Stability analysis of the proposed AGGD-ICA learning algorithm is also discussed. Compared with conventional ICA algorithm, the method can separate a wide range of source signals using only one unified density model. Simulations confirm the effectiveness and performance of the proposed algorithm.

Citation Download Citation

Fasong Wang, Hongwei Li, Rui Li, and Lihua Fu "Unified parametric ICA algorithm for hybrid sources and its stability analysis", Proc. SPIE 5985, International Conference on Space Information Technology, 59852E (4 January 2006); https://doi.org/10.1117/12.657660

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available