papers on blind source separation available in the neuroprose archive

[Kari Torkkola]

The following two related  papers on blind source separation
are available on-line in the neuroprose archive. Papers extend
the information maximization approach of Bell and Sejnowski towards 
the separation of more realistic mixtures of signals.

FTP-host: archive.cis.ohio-state.edu
FTP-file: pub/neuroprose/torkkola.icassp96.ps.Z
FTP-file: pub/neuroprose/torkkola.nnsp96.ps.Z



------------------------------------------------------------------
BLIND SEPARATION OF DELAYED SOURCES BASED ON INFORMATION MAXIMIZATION

4 pages, 341K compressed,  920K uncompressed

@InProceedings{Torkkola96icassp,
  author = 	"Kari Torkkola",
  title = 	"Blind separation of delayed sources
  		 based on information maximization",
  booktitle =	"Proceedings of the IEEE International Conference on
                 Acoustics, Speech and Signal Processing",
  address =  	"Atlanta, GA",
  month = 	"May 7-10",
  year = 	"1996",
  pages = 	"3510-3513",
  url = "ftp://archive.cis.ohio-state.edu/pub/neuroprose/torkkola.icassp96.ps.Z",
}

Abstract 

Recently, Bell and Sejnowski have presented an approach to blind source 
separation based on the information maximization principle. We extend this 
approach into more general cases where the sources may have been delayed
with respect to each other. We present a network architecture capable of 
coping with such sources, and we derive the adaptation equations for the 
delays and the weights in the network by maximizing the information 
transferred through the network.


------------------------------------------------------------------
BLIND SEPARATION OF CONVOLVED SOURCES BASED ON INFORMATION MAXIMIZATION

10 pages, 147K compressed, 340K uncompressed

@InProceedings{	Torkkola96nnsp,
  author =      "Kari Torkkola",
  title =       "Blind separation of convolved sources
  		 based on information maximization",
  booktitle =	"Neural Networks for Signal Processing VI
                 (Proceedings of the 1996 IEEE Workshop)",
  address =  	"Kyoto, Japan",
  month = 	"September 4-6",
  year = 	"1996",
  note = 	"(in press)",
  url = "ftp://archive.cis.ohio-state.edu/pub/neuroprose/torkkola.nnsp96.ps.Z",
}

Abstract 

Blind separation of independent sources from their convolutive mixtures is a 
problem in many real world multi-sensor applications. In this paper we present 
a solution to this problem based on the information maximization principle, 
which was recently proposed by Bell and Sejnowski for the case of blind 
separation of instantaneous mixtures. We present a feedback network 
architecture capable of coping with convolutive mixtures, and we derive the 
adaptation equations for the adaptive filters in the network by maximizing the 
information transferred through the network. Examples using speech signals are 
presented to illustrate the algorithm.



------------------------------------------------------------------
Kari Torkkola                          
Motorola Phoenix Corporate Research    phone: +1-602-4134129
Mail Drop EL 508                       fax:   +1-602-4137281
2100 E. Elliot Rd                      email: torkkk at base.sps.mot.com
Tempe, AZ 85284                               A540AA at email.mot.com