next up previous
Next: Linear mixture model Up: Problem Statement Previous: Problem Statement

Model for sparse sources

In order to model sources with different degrees of sparsity we consider the probability density function (pdf)

$\displaystyle p_{S_i}(s_i) = p_i\delta(s_i) + (1-p_i)f_{S_i}(s_i),\quad i=1,\dots,n;$ (1)

where $ p_i$ is the probability that a source is inactive, $ \delta(\cdot)$ denotes Dirac's delta and $ f_{S_i}(s_i)$ is the pdf of the $ i$-th source when it is active [8].



Steven Van Vaerenbergh
Last modified: 2006-04-05