online read us now
Paper details
Number 3 - September 2008
Volume 18 - 2008
Determining the weights of a Fourier series neural network on the basis of the multidimensional discrete Fourier transform
Krzysztof Halawa
Abstract
This paper presents a method for training a Fourier series neural network on the basis of the multidimensional discrete Fourier transform. The proposed method is characterized by low computational complexity. The article shows how the method can be used for modelling dynamic systems.
Keywords
orthogonal neural networks, Fourier series, fast Fourier transform, approximation, nonlinear systems