Title :
An Improved Method for the FastICA Algorithm
Author :
Zhao, Feng ; Cai, Min ; Zhang, Yunjie
Author_Institution :
Sch. of Sci., Dalian Jiaotong Univ., Dalian, China
Abstract :
The FastICA algorithm based on Newton´s iteration method can rapidly find hidden independent component from the mixed observations, and is widely used in the field of blind source separation. However, we need further improve the algorithm performance when processing massive data (such as image data). In this paper, an improved FastICA algorithm is proposed for blind source separation by establishing a Newton´s iteration method with fifth-order convergence. The simulations show that, in contrast with FastICA algorithm, proposed algorithm has comparable separation performance and fewer iteration numbers.
Keywords :
blind source separation; convergence; independent component analysis; iterative methods; FastICA algorithm; Newton iteration method; algorithm performance; blind source separation; convergence; image data; independent component analysis; separation performance; Algorithm design and analysis; Convergence; Independent component analysis; Indexes; Integrated circuits; Iterative methods; Signal processing algorithms;
Conference_Titel :
Multimedia Technology (ICMT), 2010 International Conference on
Conference_Location :
Ningbo
Print_ISBN :
978-1-4244-7871-2
DOI :
10.1109/ICMULT.2010.5630975