readme

该工具箱为时频分析中的一类重要变换,离散时频分布的工具箱,其中提供基本算法的程序.可直接应用

DiscreteTFDs

Copyright (c) 1997, 1998, 1999 Jeffrey C. O'Neill ([e-mail])
Copyright (c) 1998, 1999 Boston University
Copyright (c) 1997, 1998 L'Ecole Normale Superieure de Lyon
Copyright (c) 1997 The Regents of the University of Michigan
All Rights Reserved
see DiscreteTFDs/Copyright

This is a collection of Matlab files for time-frequency
analysis.  However, if you don't have Matlab, you can try Octave
	http://www.che.wisc.edu/octave/
or SciLab
	http://www-rocq.inria.fr/scilab/

These programs are either a result of my research or something that I
found useful enough to spend the time to implement.  Included are: a
rigorous implementation of time-frequency distributions (Cohen class),
some quartic time-frequency distributions, chirplet decomposition based
on maximum likelihood estimation, fractional Fourier transform,
time-varying filtering, and other useful little utilities.  There
isn't a manual, but you can find details in my papers
	http://mdsp.bu.edu/jeffo
or you can send me an email
	[e-mail]

The files are located in several sub-directories:

  Chirplets  -- atomic decomposition based on maximum likelihood estimation
  Demos      -- demonstrations to get you started
  Symplectic -- symplectic transformations (fractional Fourier transform)
  TypeI      -- routines for computing type I TFDs (the Cohen class)
  TypeII     -- routines for computing type II TFDs
  TypeIV     -- routines for computing type IV TFDs
  Quartic    -- some quartic functions from my research
  Utils      -- some useful utilities

You will need to add something like the follwing lines to your 
startup.m file (do not put a trailing '/' at the end of p!):

	p = '/home/jeffo/matlab/DiscreteTFDs';
	path(path, p)
	DTFDPath(p)
	clear p

If you are using a Macintosh or MS Windows, then you will need to edit 
DTFDPath.m to reflect your system path conventions.

At the MATLAB prompt type "help DiscreteTFDs" for a list of available 
functions, and use "help" to get information on each function.

For someone who knows very little about the area of TFDs and wants to
experiment, I would recommend using binomial2, born_jordan2, and
spec2 with window lengths ranging from 16 to 128.

Jeff O'Neill