Title :
Entropy of a variable slightly contaminated with another
Author_Institution :
Lab. de Modelization et Calcul, Grenoble, France
fDate :
7/1/2005 12:00:00 AM
Abstract :
This work provides the expansion up to second order of the (differential) entropy of a random variable contaminated with another small variable. It is motivated by the blind source separation problem, in which it is of interest to consider the entropy of a mixture of random variables and study how it changes with respect to the mixing proportions.
Keywords :
blind source separation; entropy; Fisher information; blind source separation problem; entropy; mixing proportion; random variable contamination; Blind source separation; Data mining; Entropy; Error correction; Mutual information; Pollution measurement; Random variables; Reactive power; Source separation; Taylor series; Blind source separation (BSS); Fisher information; entropy;
Journal_Title :
Signal Processing Letters, IEEE
DOI :
10.1109/LSP.2005.849488