Exponentially accurate approximations to periodic Lipschitz functions based on Fourier series partial sums (Q1305241)

From MaRDI portal
Revision as of 11:21, 31 January 2024 by Import240129110113 (talk | contribs) (Added link to MaRDI item.)
scientific article
Language Label Description Also known as
English
Exponentially accurate approximations to periodic Lipschitz functions based on Fourier series partial sums
scientific article

    Statements

    Exponentially accurate approximations to periodic Lipschitz functions based on Fourier series partial sums (English)
    0 references
    0 references
    0 references
    24 November 1999
    0 references
    The purpose of this interesting paper is to present an accurate and robust technique to estimate the locations of discontinuities and the associated jumps in a piecewise smooth, \(2\pi\)-periodic function \(f\), if the first \(2N+ 1\) Fourier coefficients of \(f\) are known. The method consists of two steps: 1. Find initial estimates of the locations and magnitudes of the discontinuities in \(f\) by certain properties of Gibbs phenomenon. 2. Refine these estimates by fitting the asymptotic form of the Fourier coefficients to the given coefficients using a least squares method. Then this procedure, the so-called least squares parameter estimation method, is extended to find discontinuities in \(f^{(k)}\) \((k=1,\dots, M)\). Some aspects of the robustness of this method are discussed. Further, this method is used to reconstruct the original function \(f\), as well as its first few derivatives, with exponential accuracy, if the Fourier coefficients of \(f\) are known. The technique is illustrated with several examples.
    0 references
    reconstruction of periodic
    0 references
    piecewise smooth function
    0 references
    exponentially accurate approximation
    0 references
    Fourier sum
    0 references
    numerical examples
    0 references
    locations of discontinuities
    0 references
    Gibbs phenomenon
    0 references
    Fourier coefficients
    0 references
    least-squares method
    0 references
    parameter estimation
    0 references

    Identifiers