Gabor atoms

From Wikipedia, the free encyclopedia

It has been suggested that this article or section be merged into Granular synthesis. (Discuss)

There are very few or no other articles that link to this one.
Please help introduce links in articles on related topics. After links have been created, remove this message.
This article has been tagged since September 2006.

In mathematics, Gabor atoms or Gabor functions are functions used in the analysis proposed by Dennis Gabor in 1946 where a family of functions is built from translations and modulations of a generating one.

[edit] Overview

In 1946, Dennis Gabor suggested the idea of using a granular system to produce sound. In his work, Gabor discussed the problems with the Fourier analysis, and, according to him, although the mathematics is perfectly correct, it is not possible to apply it physically, mainly in usual sounds, as the sound of a siren, in which the frequency parameter is variable through time. Another problem would be the underlying supposition, as we use sine waves analysis, that the signal under concern has infinite duration. Gabor proposes to apply the ideas from quantum physics to sound, allowing an analogy between sound and quanta. Under a mathematical background he proposed a method to reduce the Fourier analysis into cells. His research aimed the information transmission through communication channels. Gabor saw in his atoms a possibility to transmit the same information but using less data, instead of transmitting the signal itself it would be possible to transmit only the coefficients which represent the same signal using his atoms.

[edit] Mathematical Definition

The Gabor function is defined by

g l, n (x) = g (x - a l) e 2π i b n x

, $-\infty < l,n < \infty$ ,

where $a$ and $b$ are constants and $g$ is a fixed function in $L^2(\R)$ , such that $\Vert g \Vert = 1$ . It works in almost the same way as in the wavelet families, we are creating a basis for $L^2(\R)$ , using for doing so translations and modulations (instead of scaling, as done in wavelet analysis) of a base function.