Quantization Effects in the LMS and RLS Algorithms

  • Paulo Sergio Ramirez
Part of the The Kluwer International Series in Engineering and Computer Science book series (SECS, volume 694)

Abstract

In this section, several aspects of the finite-wordlength effects in the LMS algorithm are discussed for the cases of implementations in fixed- and floating-point arithmetics [1]–[3].

Keywords

Gaussian White Noise Quantization Effect Quantization Error Quantization Noise Adaptive Filter 
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.
    M. Andrews and R. Fitch, “Finite wordlength arithmetic computational error effects on the LMS adaptive weights,” Proc. IEEE Int. Conf. Acoust., Speech, and Signal Processing, pp. 628–631, May 1977.Google Scholar
  2. 2.
    C. Caraiscos and B. Liu, “A roundoff error analysis of the LMS adaptive algorithm,” IEEE Trans, on Acoust., Speech, and Signal Processing, vol. ASSP-32, pp. 34–41, Feb. 1984.Google Scholar
  3. 3.
    S. T. Alexander, “Transient weight misadjustment properties for the finite precision LMS algorithm,” IEEE Trans, on Acoust., Speech, and Signal Processing, vol. ASSP-35, pp. 1250–1258, Sept. 1987.CrossRefGoogle Scholar
  4. 4.
    A. V. Oppenheim and R. W. Schaffer, Discrete-Time Signal Processing, Prentice Hall, Englewood Cliffs, NJ, 1989.MATHGoogle Scholar
  5. 5.
    A. Antoniou, Digital Filters: Analysis, Design, and Applications, McGraw Hill, New York, NY, 2nd edition, 1993.Google Scholar
  6. 6.
    P. S. R. Diniz, E. A. B. da Silva, and S. L. Netto, Digital Signal Processing: System Analysis and Design, Cambridge University Press, Cambridge, UK, 2002.MATHCrossRefGoogle Scholar
  7. 7.
    A. B. Spirad and D. L. Snyder, “Quantization errors in floating-point arithmetic,” IEEE Trans, on Acoust., Speech, and Signal Processing, vol. ASSP-26, pp. 456–464, Oct. 1983.Google Scholar
  8. 8.
    S. H. Ardalan, “Floating-point analysis of recursive least-squares and least-mean squares adaptive filters,” IEEE Trans, on Circuits and Systems, vol. CAS-33, pp. 1192–1208, Dec. 1986.CrossRefGoogle Scholar
  9. 9.
    J. M. Cioffi, “Limited precision effects in adaptive filtering,” IEEE Trans, on Circuits and Systems, vol. CAS-34, pp. 821–833, July 1987.CrossRefGoogle Scholar
  10. 10.
    S. H. Ardalan and S. T. Alexander, “Fixed-point roundoff error analysis of the exponentially windowed RLS algorithm for time-varying systems,” IEEE Trans, on Acoust., Speech, and Signal Processing, vol. ASSP-35, pp. 770–783, June 1983.Google Scholar
  11. 11.
    G. E. Bottomley and S. T. Alexander, “A novel approach for stabilizing recursive least squares filters,” IEEE Trans, on Signal Processing, vol. 39, pp. 1770–1779, Aug. 1991.CrossRefGoogle Scholar
  12. 12.
    F. Ling and J. G. Proakis, “Numerical accuracy and stability: two problems of adaptive estimation algorithms caused by round-off error,” Proc. IEEE Intern. Conf, on Acoust., Speech, Signal Processing, San Diego, CA, pp. 30.3.1–30.3.4, March 1984.Google Scholar
  13. 13.
    G. Kubin, “Stabilization of the RLS algorithm in the absence of persistent excitation,” Proc. IEEE Intern. Conf. on Acoust., Speech, Signal Processing, NY, pp. 1369–1372, Apr. 1988.Google Scholar
  14. 14.
    M. H. Verhaegen, “Round-off error propagation in four generally applicable, recursive, least squares estimation schemes,” Automatica, vol. 25, pp. 437–444, Mar. 1989.MathSciNetMATHCrossRefGoogle Scholar
  15. 15.
    T. Adali and S. H. Ardalan, “Steady state and convergence characteristics of the fixed-point RLS algorithm,” Proc. IEEE Intern. Symp. on Circuits Systems, New Orleans, LA, pp. 788–791, May 1990.Google Scholar

Copyright information

© Springer Science+Business Media New York 2002

Authors and Affiliations

  • Paulo Sergio Ramirez
    • 1
  1. 1.Federal University of Rio de JaneiroRio de JaneiroBrazil

Personalised recommendations