A high order, progressive method for the evaluation of irregular oscillatory integrals (Q678886): Difference between revisions

From MaRDI portal
RedirectionBot (talk | contribs)
Changed an Item
Created claim: MaRDI profile type (P1460): MaRDI publication profile (Q5976449), #quickstatements; #temporary_batch_1710423558064
Property / MaRDI profile type
 
Property / MaRDI profile type: MaRDI publication profile / rank
 
Normal rank

Revision as of 16:17, 14 March 2024

scientific article
Language Label Description Also known as
English
A high order, progressive method for the evaluation of irregular oscillatory integrals
scientific article

    Statements

    A high order, progressive method for the evaluation of irregular oscillatory integrals (English)
    0 references
    0 references
    14 February 1999
    0 references
    A method for numerically evaluating integrals of irregularly oscillatory integrands of the form \(f(x)\exp(i\tau q(x))\) is developed from \textit{D. Levin's} method [Math. Comput. 38, No. 158, 531-538 (1982; Zbl 0482.65013)]. The evolution is guided by observations of the simple oscillatory case \(q(x)\equiv x\) with Chebyshev polynomials as the approximating basis. In the general case the quadrature rule is constructed to be exact for a particular set of functions suggested by joint consideration of Chebyshev polynomial approximation and Levin's method. As in the simple case the choice of nodes for the rules proves to be crucial for numerical convergence and stability. Equally-spaced points are unsuitable but Chebyshev, Clenshaw-Curtis and a similar set with only one endpoint are effective in practical cases. Computation of the weights and their required accuracy are discussed. Numerical examples are given to demonstrate the success of the method. Results are compared with alternative methods including Levin's original method.
    0 references
    Fourier integrals
    0 references
    numerical examples
    0 references
    irregular oscillatory integrals
    0 references
    quadrature rule
    0 references
    convergence
    0 references
    stability
    0 references

    Identifiers

    0 references
    0 references
    0 references
    0 references
    0 references