Rapid evaluation of nonreflecting boundary kernels for time-domain wave propagation

Bradley Alpert, Leslie Greengard, Thomas Hagstrom

Research output: Contribution to journalArticlepeer-review


We present a systematic approach to the computation of exact nonreflecting boundary conditions for the wave equation. In both two and three dimensions, the critical step in our analysis involves convolution with the inverse Laplace transform of the logarithmic derivative of a Hankel function. The main technical result in this paper is that the logarithmic derivative of the Hankel function H(1)v(z) of real order v can be approximated in the upper half z-plane with relative error ε by a rational function of degree d ∼ O(log |v| log 1/ε + log2 |v| + |v|-1 log2 1/ε) as |v| → ∞, ε → 0, with slightly more complicated bounds for v = 0. If N is the number of points used in the discretization of a cylindrical (circular) boundary in two dimensions, then, assuming that ε < 1/N, O(N log N log 1/ε) work is required at each time step. This is comparable to the work required for the Fourier transform on the boundary. In three dimensions, the cost is proportional to N2 log2 N + N2 log N log 1/ε for a spherical boundary with N2 points, the first term coming from the calculation of a spherical harmonic transform at each time step. In short, nonreflecting boundary conditions can be imposed to any desired accuracy, at a cost dominated by the interior grid work, which scales like N2 in two dimensions and N3 in three dimensions.

Original languageEnglish (US)
Pages (from-to)1138-1164
Number of pages27
JournalSIAM Journal on Numerical Analysis
Issue number4
StatePublished - 2000


  • Absorbing boundary condition
  • Approximation
  • Bessel function
  • High-order convergence
  • Maxwell's equations
  • Nonreflecting boundary condition
  • Radiation boundary condition
  • Wave equation

ASJC Scopus subject areas

  • Numerical Analysis
  • Computational Mathematics
  • Applied Mathematics


Dive into the research topics of 'Rapid evaluation of nonreflecting boundary kernels for time-domain wave propagation'. Together they form a unique fingerprint.

Cite this