Published November 1, 2013 | Version v1
Journal article

A high-order integral solver for scalar problems of diffraction by screens and apertures in three-dimensional space

Description

We present a novel methodology for the numerical solution of problems of diffraction by infinitely thin screens in three-dimensional space. Our approach relies on new integral formulations as well as associated high-order quadrature rules. The new integral formulations involve weighted versions of the classical integral operators related to the thin-screen Dirichlet and Neumann problems as well as a generalization to the open-surface problem of the classical Calderón formulae. The high-order quadrature rules we introduce for these operators, in turn, resolve the multiple Green function and edge singularities (which occur at arbitrarily close distances from each other, and which include weakly singular as well as hypersingular kernels) and thus give rise to super-algebraically fast convergence as the discretization sizes are increased. When used in conjunction with Krylov-subspace linear algebra solvers such as GMRES, the resulting solvers produce results of high accuracy in small numbers of iterations for low and high frequencies alike. We demonstrate our methodology with a variety of numerical results for screen and aperture problems at high frequencies—including simulation of classical experiments such as the diffraction by a circular disc (featuring in particular the famous Poisson spot), evaluation of interference fringes resulting from diffraction across two nearby circular apertures, as well as solution of problems of scattering by more complex geometries consisting of multiple scatterers and cavities

Availability note (English)

Available from http://dx.doi.org/10.1016/j.jcp.2013.06.022

Additional details

Identifiers

DOI
10.1016/j.jcp.2013.06.022;
arXiv
arXiv:1208.5173v1;
PII
S0021-9991(13)00446-4;

Publishing Information

Journal Title
Journal of Computational Physics
Journal Volume
252
Journal Page Range
p. 250-274
ISSN
0021-9991
CODEN
JCTPAH

Optional Information

Copyright
Copyright (c) 2013 Elsevier Science B.V., Amsterdam, The Netherlands, All rights reserved.