arXiv.itsnick.co

an independent field guide to 23 million papers from arXiv & OpenAlex — not affiliated with arXiv.org

Computer Science · Signal Processing

A method of ICA based on estimating the PDF of signals

LI Xiao-jun

See on the map

Abstract

Independent component analysis (ICA) is a method for finding independent components from multivariate (multidimensional) statistical data. Based on the optimal estimation function, a method for the estimation of the score function is developed. By using the Gaussian mixture model , an EM algorithm for approximating the probability density of the data is presented, and a stochastic gradient method is given to separate the independent components. To improve the accuracy and stability of the algorithm, an iterative method for estimating the PDF of data is presented, which can perform the separation of mixed sub-Gaussian from super-Gaussian sources. The performance of the method is shown by computer simulations.

More like this