Pearson_ICA.zip_ICA盲信号分离_pearson_极大似然ICA_盲信号分离

The Pearson-ICA package is Copyright (c) Helsinki University of Technology, Signal Processing Laboratory, Jan Eriksson, Juha Karvanen, and Visa Koivunen. This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2 of the License, or (at your option) any later version. This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details. You should have received a copy of the GNU General Public License along with this program; if not, write to the Free Software Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Version: 1.1 Version date: September 25, 2000 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Description =========== This package provides the Matlab (5.x) functions needed for the use of the Pearson-ICA algorithm as described in Karvanen, J.,Eriksson, J., and Koivunen, V.: "Pearson System Based Method for Blind Separation", Proceedings of Second International Workshop on Independent Component Analysis and Blind Signal Separation, Helsinki 2000, pp. 585--590 The algorithm is proposed to solve the standard noiseless linear ICA problem, i.e. the ICA model is Y=AS, where the number of sources s_i is equal to the number of observations y_i. %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Installation: Just put all files to a directory along Matlab's search path. ============ %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Quick use: Type pearson_ica_demo for a demonstration. ========= %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Use: === 1) Pearson-ICA algorithm ~~~~~~~~~~~~~~~~~~~~~~~~ Suppose you have the ICA mixture in the matrix `mixedsig', where the different rows correspond to the different outputs. Then estimatedsig=pearson_ica(mixedsig) gives the estimated independent components as the rows of the matrix `estimatedsig'. The estimated mixing matrix A and estimated separation matrix W are obtained as [estimatedsig,A,W]=pearson_ica(mixedsig) There are also some optional parameters you can change: 'epsilon' Convergence criterion 'maxNumIterations' Maximum number of iterations 'borderBase', 'borderSlope' The border lines between the Pearson family and the tanh contrast. I.e. the Pearson is used if borderBase(1)+borderSlope(1)*skewness^2=< kurtosis=<borderBase(2)+borderSlope(2)*skewness^2, and the contrast tanh is used otherwise. The default values for the parameters are epsilon=0.0001, maxNumIterations=200, borderBase=[2.6 4], and borderSlope=[0 1]. Example: estimated_sig=pearson_ica(mixedsig,'epsilon',0.00005,... 'maxNumIterations',50,'borderBase',[2.5 4]) 2) Use of the functions and the list of files in the package ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ contents.m A brief description of all the files in the package. fasticapearson.m Strip down of Hyv�rinen's FastICA algorithm. This is the core function used by pearson_ica. Use pearson_ica to call this function. Syntax: [ICs, A, W]=fasticapearson(mixedsig,epsilon,... maxNumIterations,borderBase,borderSlope); Dependences: gbd_momentfit, gbd_score, pearson_momentfit, pearson_score gbd_momentfit.m Provides the gbd_momentfit function, which estimates the parameters of a distribution from the Generalized Beta Distribution (GBD) using the first four sample moments. Alternatively, the sample minimum and maximum can be used instead of the sample mean and variance. Syntax: gbd_momentfit(alpha3,alpha4,samplemin,samplemax,samplen) gbd_momentfit(alpha3,alpha4,alpha1,alpha2) Examples: beta=gbd_momentfit(0.254,2.526,57,580) beta=gbd_momentfit(0.254,2.526,-0.68,144,1000) gbd_score.m Provides the gbd_score function, which calculates values of the score function and its derivative of a distribution from the Generalized Beta Distribution (GBD). Syntax: [phi,phi_prime]=gbd_score(x,beta) Example: x=0:0.1:140; beta=gbd_momentfit(0.254,2.526,57,580); [phi,phi2]=gbd_score(x,beta); plot(x,phi); gpl.txt The GNU General Public License pearson_ica.m The main file for the Pearson-ICA algorithm. Look above for instructions how to call this function. pearson_ica_demo.m A demonstration of the Pearson-ICA algorithm in work. pearson_momentfit.m Estimates the parameters of the zero mean and unit variance Pearson system using the third and forth central moments. Syntax: pearson_momentfit(alpha3,alpha4) Example: b_normal=pearson_momentfit(0,3) pearson_score.m Calculates values for the score function and its derivative of the Pearson system with parameters given in B=[b(1) b(2) b(3)]. Syntax: [phi,phi_prime]=pearson_score(x,B) Example: x=-3:0.01:3; b_normal=pearson_momentfit(0,3); [phi,phi_prime]=pearson_score(x,b_normal); plot(x,phi_prime); readme.txt This file. % The end of readme.txt %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%