Advertisement

Neural Networks for Higher-Order Spectral Estimation

  • F.-L. Luo
  • R. Unbehauen
Conference paper

Abstract

This paper deals with neural network approaches for higher order spectral estimation. The emphasis is put on how to use analog neural networks to perform in realtime major computations required in the ARMA model based bispectral estimation and the fourth order cumulant based Pisarenko’s harmonic method. The proposed approaches are useful for the real-time signal processing with higher order spectral estimation.

Keywords

Neural Network ARMA Model Neural Network Approach High Order Spectrum Nonminimum Phase 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. [1]
    G. B. Giannakis. On the identifiability of non-Gaussian ARMA models using cumulants. IEEE Trans, on Automatic Control, 35(1):18–26, 1990.MathSciNetMATHCrossRefGoogle Scholar
  2. [2]
    G. B. Giannakis and A. Swami. On estimating non-causal nonminimum phase ARMA models of non-Gaussian processes. IEEE Trans, on Acoustic Speech and Signal Processing, 38(3):478–495, 1990.MathSciNetMATHCrossRefGoogle Scholar
  3. [3]
    F. L. Luo and R. Unbehauen. Neural network approach to Pisarenko’s frequency estimation. In Proceedings of ICANNGA’ 95, pages 261–264, 1995.Google Scholar
  4. [4]
    F. L. Luo and R. Unbehauen. Applied Neural Networks for Signal Processing. Cambridge University Press, New York, 1997.Google Scholar
  5. [5]
    C. L. Nikias and J. M. Mendel. Signal processing with higher-order spectral. IEEE Signal Processing Magazine, 10:10–37, July 1993.CrossRefGoogle Scholar
  6. [6]
    A. Swami and J. M. Mendel. ARMA parameter estimation using only output cumulants. IEEE Trans, on Acoustic Speech and Signal Processing, 38(7):1257–1265, 1990.MathSciNetCrossRefGoogle Scholar
  7. [7]
    D. W. Tank and J. J. Hopfield. Simple ‘neural’ optimization networks: A/D converter, signal decision circuit, and linear programming circuit. IEEE Trans, on Circuits and Systems, 33:533–541, 1986.CrossRefGoogle Scholar
  8. [8]
    X. D. Zhang and Y. L. Zhou. A novel recursive approach to estimation MA parameters of casual ARMA models from cumulants. IEEE Trans, on Signal Processing, 40(1):2870–2873, 1992.CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Wien 1998

Authors and Affiliations

  • F.-L. Luo
    • 1
  • R. Unbehauen
    • 1
  1. 1.Lehrstuhl für Allgemeine und Theoretische ElektrotechnikUniversität Erlangen-NürnbergErlangenGermany

Personalised recommendations