Software for time-frequency analysis. For background on the subject, see: this page .
Matlab and C language programs for computing time-frequency representations (TFRs):
* OPTIMAL RADIALLY GAUSSIAN KERNEL TFR
Batch oriented optimal-kernel TFR. Computes one optimal kernel for the entire signal. (Matlab)
Download the most recent source code (Version 1.0, February 1996) of rgk.m or view the documentation and installation information. (The tar file archive contains both latter files.)
This software is distributed "as-is" and is free of charge. For further information on the conditions of use please read the following message.
* ADAPTIVE OPTIMAL-KERNEL TFR (AOK) TFR
Online optimal-kernel TFR suitable for long signals with changing time-frequency characteristics. Adapts the kernel over time for improved performance. (Matlab-compatible C)
C language source code implementation of the algorithm derived in
* D. L. Jones and R. G. Baraniuk, An Adaptive Optimal-Kernel Time-Frequency Representation, IEEE Transactions on Signal Processing, vol. 43, no. 11, pp. 2361-2371, October 1995. Abstract.
Program outputs Matlab-compatible ASCII data file for ease of display. A C-Mex version for Matlab is planned for the year 2040.
Example: AOK TFR of a train of bottlenose dolphin clicks
Download the most recent source code (Version 4.0, February 1996) of aok4.c or view the documentation and installation information. (The tar file archive contains both latter files.)
This software is distributed as-is and free of charge. For further information on the conditions of use please read the README in the software package.
PC/Mac compatible versions of the AOK software from Erik Winkler.
A flexible class of time-scale covariant TFRs. Software is available as part of the Matlab time-frequency toolbox.
* TIME-FREQUENCY TOOLBOX
A library of Matlab functions for time-frequency analysis.
For more info see Optimal-Kernel Time-Frequency Representations.