readme

一个使用空间时频分布进行盲信号分离的Matlab程序

******************************* The TFBSS code *************************************

TFBSS performs Blind Source Separation of (over)determined multiplicative mixtures
of non-stationary real valued sources.

TFBSS can be downloaded at http://www.irccyn.ec-nantes.fr/~fevotte/TFBSS_pack

TFBSS is based on the joint-diagonalization of whitened and noise-compensated
Spatial Time-Frequency Distribution (STFD) matrices of the observations,
corresponding to single auto-terms positions.

------------------------------------------------------------------------------------
The current main reference  is:

A. Holobar, C. F関otte, C. Doncarli, and D. Zazula,
"Single autoterms selection for blind source separation in time-frequency plane",
In 11 e EUSIPCO, Toulouse, France, 3-6 septembre 2002.

The iterative selection of maxima of the criteria in the above paper has been 
replaced by a more simple and computation friendly gradient approach to be published
soon. The inner Iterative Joint Diagonalization has not been implemented in TFBSS.

The first paper dealing with joint-diagonalization of STFD matrices is:

A. Belouchrani and M. G. Amin,
"Blind Source Separation Based on Time-Frequency Signal Representation'',
IEEE Trans. on Signal Processing, vol. 46. No. 11. pp. 2888-2898. November 1998.
------------------------------------------------------------------------------------

The TFBSS pack contains the following MATLAB files:

* tfbss.m : main program

* joint_diag.m : performs joint-diagonalization of complex matrices. Available on 
J.F Cardoso web site (http://www.tsi.enst.fr/~cardoso/stuff.html) and used with
the kind permission of its author. 
Copyright owned by J.F Cardoso - [e-mail] .

* tfrridb.m & window.m : perform TFDs computation. Excerpts from the Matlab 
Time-Frequency Toolbox and used with the kind permission of their author F. Auger. 
Copyright owned by F. Auger - [e-mail] . 
Download the whole Matlab Time-Frequency Toolbox at:
http://crttsn.univ-nantes.fr/~auger/tftb.html

* demoTFBSS.m : demo script of TFBSS, separation of 4 noisy instantaneous mixtures of
3 Time-Varying ARMA sources.

* data.mat : contains the sources used in demoTFBSS.

For a quick overview of TFBSS performance, download all the files in a common 
directory and call demoTFBSS.m in a MATLAB command window.

Feedback about your tests with real world signals is welcome.
Please mail me results.

Please report any bug.

Author: C. F関otte ([e-mail])
Copyright is owned by C. F関otte & A. Holobar

-------------------------------------------------------------------------------------
TFBSS.m - Sept 9, 2002