• DocumentCode
    1098752
  • Title

    Bias and variance of spectral estimates from an all-pole digital filter

  • Author

    Davis, B.R. ; Cowley, W.G.

  • Author_Institution
    University of Adelaide, Adelaide, South Australia
  • Volume
    30
  • Issue
    2
  • fYear
    1982
  • fDate
    4/1/1982 12:00:00 AM
  • Firstpage
    322
  • Lastpage
    329
  • Abstract
    An adaptive digital filter, equivalent to a one step linear predictor, may be employed to determine spectral estimates closely related to the linear prediction all-pole model spectrum of the input signal. This paper presents an analysis of the bias and variance of this estimate for a filter using the Widrow-Hoff LMS algorithm. Results are compared with computer simulations for two four pole signals.
  • Keywords
    Adaptive filters; Algorithm design and analysis; Analysis of variance; Convergence; Digital filters; Eigenvalues and eigenfunctions; Equations; Least squares approximation; Predictive models; Signal processing algorithms;
  • fLanguage
    English
  • Journal_Title
    Acoustics, Speech and Signal Processing, IEEE Transactions on
  • Publisher
    ieee
  • ISSN
    0096-3518
  • Type

    jour

  • DOI
    10.1109/TASSP.1982.1163884
  • Filename
    1163884
    • Link To Document
    https://search.isc.ac/dl/search/defaultta.aspx?DTC=49&DC=1098752