الفهرس 
IntroductionOnly 14 pages are availabe for public view
 |  | from | 132 |  |  |
| | | from | 132 | | |
Abstract
This thesis addresses the problem of blind source separation and multichannel blind deconvolution . In blind source separation, signals from multiple sources arrive simultaneously at a sensor array, so, each sensor output contains a mixture of source signals. Sets of sensor outputs are processed to recover the source signals from the mixed observations. A multichannel blind deconvolution can be considered as a natural extension or generalization of the instantaneous blind separation problem. The goal of Multichannel deconvolution is to calculate the possibly scaled and time delayed (or filtered) versions of the source signals from the received signals by using approximate knowledge of the source signal distributions and statistics. Application domains for the material in this thesis include communications, biomedical, and sensor array signal processing.
The goal of this thesis is to develop a blind source recovery algorithms for a mixture of Sub and Super gaussian signals in case of the model represented by state space equations. Then The problem of optimal choice of nonlinear activation functions for various distributions, e.g., Gaussian, Laplacian, impulsive and uniformly-distributed signals based on a Generalized-Gaussian-distributed model is considered . Furthermore, a rich family of distributions based on Generalized gamma , Generalized Logistic and Weibull are also proposed as probability density functions for signals of sub and super Gaussian distributed models. The proposed parametric score functions are derived from the generalized gamma , generalized Gaussian, Generalized Logistic and Weibull distribution models. An adaptive algorithms to determine the parameters for the proposed score function using mutual information of Blind Source Recovery(BSR) output are also presented. Primary advantage of the proposed frameworks are that they render the adaptive estimation of the demixing network to be completely blind. No a priori information about the distribution structure of the original sources is required.