Projection methods for improved performance in FIR adaptive filters

R. A. Soni, K. A. Gallivan, W. K. Jenkins

    Research output: Contribution to conferencePaper

    3 Citations (Scopus)

    Abstract

    The normalized LMS algorithms offer low-computational complexity and inexpensive implementations for FIR adaptive filters. However, convergence rate decreases as the eigenvalue ratio (condition number) of the input autocorrelation matrix increases. Recursive least squares methods offer significant convergence rate improvement but at the expense of increased computational complexity. In this paper, we present a class of algorithms, collectively called Projection Methods, which offers flexibility in the tradeoff between computational complexity and convergence rate improvement. These methods are related to traditional normalized data reusing algorithms described by Schnaufer and Jenkins. Utilizing conjugate gradient and Tchebyshev methods, algorithms are developed which accelerate the convergence behavior of traditional normalized data reusing algorithms while maintaining excellent tracking performance.

    Original languageEnglish (US)
    Pages746-749
    Number of pages4
    StatePublished - Dec 1 1997
    EventProceedings of the 1997 40th Midwest Symposium on Circuits and Systems. Part 1 (of 2) - Sacramento, CA, USA
    Duration: Aug 3 1997Aug 6 1997

    Other

    OtherProceedings of the 1997 40th Midwest Symposium on Circuits and Systems. Part 1 (of 2)
    CitySacramento, CA, USA
    Period8/3/978/6/97

    Fingerprint

    FIR filters
    Adaptive filters
    Computational complexity
    Autocorrelation

    All Science Journal Classification (ASJC) codes

    • Electronic, Optical and Magnetic Materials
    • Electrical and Electronic Engineering

    Cite this

    Soni, R. A., Gallivan, K. A., & Jenkins, W. K. (1997). Projection methods for improved performance in FIR adaptive filters. 746-749. Paper presented at Proceedings of the 1997 40th Midwest Symposium on Circuits and Systems. Part 1 (of 2), Sacramento, CA, USA, .
    Soni, R. A. ; Gallivan, K. A. ; Jenkins, W. K. / Projection methods for improved performance in FIR adaptive filters. Paper presented at Proceedings of the 1997 40th Midwest Symposium on Circuits and Systems. Part 1 (of 2), Sacramento, CA, USA, .4 p.
    @conference{e4748d6dede04c55b9104628e1860277,
    title = "Projection methods for improved performance in FIR adaptive filters",
    abstract = "The normalized LMS algorithms offer low-computational complexity and inexpensive implementations for FIR adaptive filters. However, convergence rate decreases as the eigenvalue ratio (condition number) of the input autocorrelation matrix increases. Recursive least squares methods offer significant convergence rate improvement but at the expense of increased computational complexity. In this paper, we present a class of algorithms, collectively called Projection Methods, which offers flexibility in the tradeoff between computational complexity and convergence rate improvement. These methods are related to traditional normalized data reusing algorithms described by Schnaufer and Jenkins. Utilizing conjugate gradient and Tchebyshev methods, algorithms are developed which accelerate the convergence behavior of traditional normalized data reusing algorithms while maintaining excellent tracking performance.",
    author = "Soni, {R. A.} and Gallivan, {K. A.} and Jenkins, {W. K.}",
    year = "1997",
    month = "12",
    day = "1",
    language = "English (US)",
    pages = "746--749",
    note = "Proceedings of the 1997 40th Midwest Symposium on Circuits and Systems. Part 1 (of 2) ; Conference date: 03-08-1997 Through 06-08-1997",

    }

    Soni, RA, Gallivan, KA & Jenkins, WK 1997, 'Projection methods for improved performance in FIR adaptive filters', Paper presented at Proceedings of the 1997 40th Midwest Symposium on Circuits and Systems. Part 1 (of 2), Sacramento, CA, USA, 8/3/97 - 8/6/97 pp. 746-749.

    Projection methods for improved performance in FIR adaptive filters. / Soni, R. A.; Gallivan, K. A.; Jenkins, W. K.

    1997. 746-749 Paper presented at Proceedings of the 1997 40th Midwest Symposium on Circuits and Systems. Part 1 (of 2), Sacramento, CA, USA, .

    Research output: Contribution to conferencePaper

    TY - CONF

    T1 - Projection methods for improved performance in FIR adaptive filters

    AU - Soni, R. A.

    AU - Gallivan, K. A.

    AU - Jenkins, W. K.

    PY - 1997/12/1

    Y1 - 1997/12/1

    N2 - The normalized LMS algorithms offer low-computational complexity and inexpensive implementations for FIR adaptive filters. However, convergence rate decreases as the eigenvalue ratio (condition number) of the input autocorrelation matrix increases. Recursive least squares methods offer significant convergence rate improvement but at the expense of increased computational complexity. In this paper, we present a class of algorithms, collectively called Projection Methods, which offers flexibility in the tradeoff between computational complexity and convergence rate improvement. These methods are related to traditional normalized data reusing algorithms described by Schnaufer and Jenkins. Utilizing conjugate gradient and Tchebyshev methods, algorithms are developed which accelerate the convergence behavior of traditional normalized data reusing algorithms while maintaining excellent tracking performance.

    AB - The normalized LMS algorithms offer low-computational complexity and inexpensive implementations for FIR adaptive filters. However, convergence rate decreases as the eigenvalue ratio (condition number) of the input autocorrelation matrix increases. Recursive least squares methods offer significant convergence rate improvement but at the expense of increased computational complexity. In this paper, we present a class of algorithms, collectively called Projection Methods, which offers flexibility in the tradeoff between computational complexity and convergence rate improvement. These methods are related to traditional normalized data reusing algorithms described by Schnaufer and Jenkins. Utilizing conjugate gradient and Tchebyshev methods, algorithms are developed which accelerate the convergence behavior of traditional normalized data reusing algorithms while maintaining excellent tracking performance.

    UR - http://www.scopus.com/inward/record.url?scp=0031380396&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=0031380396&partnerID=8YFLogxK

    M3 - Paper

    AN - SCOPUS:0031380396

    SP - 746

    EP - 749

    ER -

    Soni RA, Gallivan KA, Jenkins WK. Projection methods for improved performance in FIR adaptive filters. 1997. Paper presented at Proceedings of the 1997 40th Midwest Symposium on Circuits and Systems. Part 1 (of 2), Sacramento, CA, USA, .