1 / 10
文档名称:

A feedforward neural network for the wavelet decomposition of discrete time signals - Neural Networks for Processing [1993] III. Proceedings of the 1993 IEEE-SP Workshop.pdf

格式:pdf   页数:10
下载后只包含 1 个 PDF 格式的文档,没有任何的图纸或源代码,查看文件列表

如果您已付费下载过本站文档,您可以点这里二次下载

A feedforward neural network for the wavelet decomposition of discrete time signals - Neural Networks for Processing [1993] III. Proceedings of the 1993 IEEE-SP Workshop.pdf

上传人:bolee65 2014/1/23 文件大小:0 KB

下载得到文件列表

A feedforward neural network for the wavelet decomposition of discrete time signals - Neural Networks for Processing [1993] III. Proceedings of the 1993 IEEE-SP Workshop.pdf

文档介绍

文档介绍:A FEEDFORWARD WORK
F'OR THE WAVELET POSITION
OF DISCRETE TIME SIGNALS
Sylvie Marcos and Messaoucl Benitlir
Laboratoire des Signaux et SystPmes, .
Plateau de Moulon, 91 192 (iifsur-Yvette, France
Phone : (33) 1 69 41 $0 40, Fax : (33) 1 69 41 30 60
E-mail : Marcos(
Abstract - In this paper a feedforward work with sigrnoYdal
activation functions is proposed to perform the wavelet po-
sition of a discrete time signals.
1 INTRODUCTION
The problem of representing a signal as an expansion of elementary functions
is very important in signal processing. The aim of such a representation is
often to reveal temporal and frequential properties of a signal in order to
characterize and process this signal. For example, the Fourier expansion of
a signal with respect to a trigonometric system has been widely used for the
frequency analysis of a signal. The distribution of coefficients appearing in
the Fourier expansion provides information about the frequency
of the original signal. However, in a non stationary context, the Fourier ex-
pansion fails in providing information. It is then useful to be able to obtain
an expansion of a signal in terms of elementary functions are well IC-
calized both in time and frequency. The wavelet expansion of a signal has
appeared to be a good wa