Stability and error estimates for Filon-Clenshaw-Curtis rules for highly-oscillatory integrals

Victor Dominguez, Ivan Graham, Valery Smyshlyaev

Research output: Contribution to journalArticle

  • 50 Citations

Abstract

In this paper we obtain new results on Filon-type methods for computing oscillatory integrals of the form $\int_{-1}^1 f(s) \exp({\rm i}ks) \ {\rm d}s $. We use a Filon approach based on interpolating $f$ at the classical Clenshaw-Curtis points $\cos(j\pi/N), \ j = 0, \ldots , N$. The rule may be implemented in $\mathcal{O}(N \log N)$ operations. We prove error estimates which show explicitly how the error depends both on the parameters $k$ and $N$ and on the Sobolev regularity of $f$. In particular we identify the regularity of $ f$ required to ensure the maximum rate of decay of the error as $k \rightarrow \infty$. We also describe a method for implementating the method and prove its stability both when $N \leq k$ and $N>k$. Numerical experiments illustrate both the stability of the algorithm and the sharpness of the error estimates.
LanguageEnglish
Pages1253-1280
Number of pages28
JournalIMA Journal of Numerical Analysis
Volume31
Issue number4
DOIs
StatusPublished - 2011

Fingerprint

Oscillatory Integrals
Stability Estimates
Error Estimates
Regularity
Sharpness
Pi
Numerical Experiment
Decay
Computing
Experiments

Cite this

Stability and error estimates for Filon-Clenshaw-Curtis rules for highly-oscillatory integrals. / Dominguez, Victor; Graham, Ivan; Smyshlyaev, Valery.

In: IMA Journal of Numerical Analysis, Vol. 31, No. 4, 2011, p. 1253-1280.

Research output: Contribution to journalArticle

@article{bbfd92838fb44bd1a2265b1d4b7aec3a,
title = "Stability and error estimates for Filon-Clenshaw-Curtis rules for highly-oscillatory integrals",
abstract = "In this paper we obtain new results on Filon-type methods for computing oscillatory integrals of the form $\int_{-1}^1 f(s) \exp({\rm i}ks) \ {\rm d}s $. We use a Filon approach based on interpolating $f$ at the classical Clenshaw-Curtis points $\cos(j\pi/N), \ j = 0, \ldots , N$. The rule may be implemented in $\mathcal{O}(N \log N)$ operations. We prove error estimates which show explicitly how the error depends both on the parameters $k$ and $N$ and on the Sobolev regularity of $f$. In particular we identify the regularity of $ f$ required to ensure the maximum rate of decay of the error as $k \rightarrow \infty$. We also describe a method for implementating the method and prove its stability both when $N \leq k$ and $N>k$. Numerical experiments illustrate both the stability of the algorithm and the sharpness of the error estimates.",
author = "Victor Dominguez and Ivan Graham and Valery Smyshlyaev",
year = "2011",
doi = "10.1093/imanum/drq036",
language = "English",
volume = "31",
pages = "1253--1280",
journal = "IMA Journal of Numerical Analysis",
issn = "0272-4979",
publisher = "Oxford University Press",
number = "4",

}

TY - JOUR

T1 - Stability and error estimates for Filon-Clenshaw-Curtis rules for highly-oscillatory integrals

AU - Dominguez, Victor

AU - Graham, Ivan

AU - Smyshlyaev, Valery

PY - 2011

Y1 - 2011

N2 - In this paper we obtain new results on Filon-type methods for computing oscillatory integrals of the form $\int_{-1}^1 f(s) \exp({\rm i}ks) \ {\rm d}s $. We use a Filon approach based on interpolating $f$ at the classical Clenshaw-Curtis points $\cos(j\pi/N), \ j = 0, \ldots , N$. The rule may be implemented in $\mathcal{O}(N \log N)$ operations. We prove error estimates which show explicitly how the error depends both on the parameters $k$ and $N$ and on the Sobolev regularity of $f$. In particular we identify the regularity of $ f$ required to ensure the maximum rate of decay of the error as $k \rightarrow \infty$. We also describe a method for implementating the method and prove its stability both when $N \leq k$ and $N>k$. Numerical experiments illustrate both the stability of the algorithm and the sharpness of the error estimates.

AB - In this paper we obtain new results on Filon-type methods for computing oscillatory integrals of the form $\int_{-1}^1 f(s) \exp({\rm i}ks) \ {\rm d}s $. We use a Filon approach based on interpolating $f$ at the classical Clenshaw-Curtis points $\cos(j\pi/N), \ j = 0, \ldots , N$. The rule may be implemented in $\mathcal{O}(N \log N)$ operations. We prove error estimates which show explicitly how the error depends both on the parameters $k$ and $N$ and on the Sobolev regularity of $f$. In particular we identify the regularity of $ f$ required to ensure the maximum rate of decay of the error as $k \rightarrow \infty$. We also describe a method for implementating the method and prove its stability both when $N \leq k$ and $N>k$. Numerical experiments illustrate both the stability of the algorithm and the sharpness of the error estimates.

UR - http://www.scopus.com/inward/record.url?scp=80054705865&partnerID=8YFLogxK

UR - http://dx.doi.org/10.1093/imanum/drq036

U2 - 10.1093/imanum/drq036

DO - 10.1093/imanum/drq036

M3 - Article

VL - 31

SP - 1253

EP - 1280

JO - IMA Journal of Numerical Analysis

T2 - IMA Journal of Numerical Analysis

JF - IMA Journal of Numerical Analysis

SN - 0272-4979

IS - 4

ER -