Papers on neural and neuromorphic blind signal processing.

[Simone G.O. Fiori (Pg)]

Dear Connectionists,
 I would like to draw your attention to three new papers of potential 
interest to people working on unsupervised artificial neural networks 
and neuromorphic adaptive filtering for blind signal processing.

 Best regards,
 Simone Fiori


        Hybrid Independent Component Analysis by Adaptive LUT 
                    Activation Function Neurons
 ==================================================================
 by  S. Fiori - Neural Networks and Adaptive Systems Research Group
     University of Perugia, Perugia (Italy)

 Journal: Neural Networks (Pergamon press)
 Download at: http://www.unipg.it/~sfr/publications/nnt2001.ps

                           Abstract

The aim of this paper is to present an efficient implementation of
unsupervised adaptive-activation function neurons dedicated to 
one-dimensional probability density estimation, with application
to independent component analysis. The proposed implementation
is a computationally light improvement to adaptive pseudo-polynomial 
neurons, recently presented in (Fiori, 2000a), and bases upon the 
concept of `look-up table' (LUT) neurons.

Keywords: Adaptive activation function neurons; Look-up-table neurons; 
Independent component analysis; Minimal mutual information principle; 
Kullback-Leibler divergence; Natural gradient.

                      ~~~~~~~~~~~~~~~~~~~~

   Notes on Cost Functions and Estimators for `Bussgang' Adaptive 
                        Blind Equalization
 ==================================================================
 by  S. Fiori - Neural Networks and Adaptive Systems Research Group
     University of Perugia, Perugia (Italy)

 Journal: European Transactions on Telecommunications (ETT)
 Download at: http://www.unipg.it/~sfr/publications/ett2001.ps

                              Abstract

The aim of this paper is to present some remarks on the cost functions 
for blind channel equalization by `Bussgang' algorithms recently discused
in the literature; also, some possible associated non-Bayesian estimators 
are considered with details, and the effects of the choice of such 
estimators in relation to `Bussgang' neuromorphic filtering are briefly
investigated.

                        ~~~~~~~~~~~~~~~~~~~~

 A Contribution to (Neuromorphic) Blind Deconvolution by Flexible 
                 Approximated Bayesian Estimation
 =================================================================
 by  S. Fiori - Neural Networks and Adaptive Systems Research Group
     University of Perugia, Perugia (Italy)

Journal: Signal Processing (Elsevier)
Download at: http://www.unipg.it/~sfr/publications/SIGPROC2001.ps

                             Abstract

`Bussgang' deconvolution techniques for blind digital channels 
equalization rely on a Bayesian estimator of the source sequence defined 
on the basis of channel/equalizer cascade model which involves the 
definition of deconvolution noise. In this paper we consider four 
`Bussgang' blind deconvolution algorithms for uniformly-distributed 
source signals and investigate their numerical performances as well as 
some of their analytical features. Particularly, we show that the algorithm, 
introduced by the present author, provided by a flexible (neuromorphic) 
estimator is effective as it does not require to make any hypothesis about 
convolutional noise level and exhibits satisfactory numerical performances.
===================================================
Dr Simone Fiori (EE, PhD)- Assistant Professor
Neural Networks and Adaptive Systems Research Group
DIE - University of Perugia - Perugia (Italy)
eMail: sfr at unipg.it Fax: +39 0744 492925
Web: http://www.unipg.it/~sfr/
===================================================