Improved Algorithm for Independent Component Analysis

JI Ce,YU Yang,YU Peng
DOI: https://doi.org/10.1109/icfcc.2010.5497485
2010-01-01
Abstract:Firstly, this paper introduces the basic theory of Independent Component Analysis (ICA) and the FastICA algorithm. In order to increase the algorithm convergence rate and reduce the running time, the paper amends the Newton's iteration method and gives an improved algorithm of independent component analysis-the Newton's iteration method with fifth-order convergence. The simulation result of the image signal separation shows that the improved algorithm have the same separate effect as conventional FastICA algorithm and can reduce the times of iterations and the running time significantly. Furthermore, the improved algorithm can also increase the convergence rate and operation efficiency.
What problem does this paper attempt to address?