Source separation of astrophysical images using particle filters: Matlab Codes - Version 1.0

This document describes the Matlab codes used to perform the simulations presented in [2]. Our aim is to separate a superposition of astrophysical images: most of the previous work on the problem perform a blind separation, assume noiseless models, and in the few cases when noise is taken into account it is generally assumed to be Gaussian and space-invariant. The code presented here has been implemented for the non-blind solution of the source separation problem using the approach named particle filtering. This method is an advanced Bayesian estimation technique which can deal with non-Gaussian and nonlinear models, and additive space-varying noise, in the sense that it is a generalization of the Kalman Filter. In this work, particle filters are utilized with objectives of both noise altering and separation of signals: this approach is extremely flexible, as it is possible to exploit the available a-priori information about the statistical properties of the sources through the Bayesian theory. The codes presented here have been developed by the author himself, as an extension of a more general-purpose code written by Alijah Ahmed, in order to deal with the astrophysical context presented in [2]. Especially in case of low SNR, the simulations show that the output quality of the separated signals is better than that of ICA, which is one of the most widespread methods for source separation. On the other hand, since a wide set of parameters, which can take from a large range of values, has to be initialized, the use of this approach needs extensive experimentation and testing.