Generalized Spectrum Analysis by Means of Neural Networks
The problem of determination of generalized signal spectrum using neural networks has been investigated. The paper contains the formalization of the problem and is focused on building of a neural network that may be used to calculate coefficients of the generalized Fourier series for various base functions. The theoretic considerations have been illustrated by an example of analysis of a signal belonging to the Hilbert space of signals with finite energy for Laguerre’s base and a wavelet base.
KeywordsNeural Network Hilbert Space Orthogonal Polynomial Recurrent Neural Network Wavelet Base
Unable to display preview. Download preview PDF.
- 1.Ayer S, Schroeter P, Brigger P (1994) Time-varying motion estimation using orthogonal polynomials and applications. Proceedings of 12-th International Conference on Pattern Recognition, Jerusalem, Israel, pp 409–414Google Scholar
- 4.Harmuth HF (1992) Transmission of information by orthogonal functions. Springer-Verlag, New YorkGoogle Scholar
- 5.Juszczyk W, Zajac M (1994) Fault diagnosis methods in selected synchronous drive system. Proceedings of 9-th Symposium on Basic Problems in Power Electronics and Electromechanics, Wisla, Poland, pp 170–175 (in Polish)Google Scholar
- 6.Mandziuk J (2000) Hopfield neural networks. Exit, Warsaw (in Polish)Google Scholar
- 7.Philips W, Jonghe G (1992) Data compression of ECGs by high degree polynomial approximation. IEEE Trans. Biomedical Eng. 39: 330–337Google Scholar
- 8.Popularkis AD (1996) The transforms and applications handbook. CRC and IEEE Press, Boca Raton, FloridaGoogle Scholar
- 9.Soliman SS, Srinath MD (1998) Continuous and discrete signals and systems. Prentice-Hall, Upper Saddle River, New JerseyGoogle Scholar
- Yaroslaysky L, Eden H (1996) Fundamentals of digital optics. Digital signal processing in optics and holography. Birkhauser, BostonGoogle Scholar