|
Title:
|
Efficient algorithms for discrete time-frequency distributions |
|
Author:
|
O'Toole, J; Mesbah, M; Boashash, B
|
|
Abstract:
|
Time-frequency distributions (TFDs) are computationally costly to compute. This paper presents algorithms to reduce this computational load. Before we can compute the TFDs, however, we must define a discrete version of the TFD. Defining a discrete TFD (DTFD) is not a straightforward process---for example, a popular DTFD definition does not satisfy all desirable mathematical properties that are inherent to the continuous TFD. In this paper, we define a new DTFD definition, the DTFD-C. This definition is closely related to another DTFD definition which we recently proposed, the DTFD-B. The DTFD-B and DTFD-C satisfy all desirable properties. We provide algorithms for both these definitions and show that the DTFD-C requires only 50% of the computational load required to compute the DTFD-B. |
|
Description:
|
This paper shows that a new discrete time version of TFDs the DTFD-C, has the clear advantage over the other definitions as it provides the signal analyst with a high performance tool, by retaining all useful properties, at a low computational cost.
(Additional details can be found in the comprehensive book on Time-Frequency Signal Analysis and Processing (see http://www.elsevier.com/locate/isbn/0080443354).
In addition, the most recent upgrade of the original software package that calculates Time-Frequency Distributions and Instantaneous Frequency estimators can be downloaded from the web site: www.time-frequency.net. This was the first software developed in the field, and it was first released publicly in 1987 at the 1st ISSPA conference held in Brisbane, Australia, and then continuously updated). |
|
URI:
|
http://hdl.handle.net/10576/10804
|
|
Date:
|
2008 |