TY - JOUR
T1 - Fast Computation of High-Frequency Dirichlet Eigenmodes via Spectral Flow of the Interior Neumann-to-Dirichlet Map
AU - Barnett, Alex
AU - Hassell, Andrew
PY - 2014/3
Y1 - 2014/3
N2 - We present a new algorithm for numerical computation of large eigenvalues and associated eigenfunctions of the Dirichlet Laplacian in a smooth, star-shaped domain in ℝd, d≥2. Conventional boundary-based methods require a root search in eigenfrequency k, hence take O(N3) effort per eigenpair found, where N=O(kd-1) is the number of unknowns required to discretize the boundary. Our method is O(N) faster, achieved by linearizing with respect to k the spectrum of a weighted interior Neumann-to-Dirichlet (NtD) operator for the Helmholtz equation. Approximations k^j to the square roots kj of all O(N) eigenvalues lying in [k-ε{lunate}, k], where ε{lunate}=O(1), are found with O(N3) effort. We prove an error estimate |k^j-kj|≤C(∈2k+∈3), with C independent of k. We present a higher-order variant with eigenvalue error scaling empirically as O(ε{lunate}5) and eigenfunction error as O(ε{lunate}3), the former improving upon the "scaling method" of Vergini and Saraceno. For planar domains (d=2), with an assumption of absence of spectral concentration, we also prove rigorous error bounds that are close to those numerically observed. For d=2 we compute robustly the spectrum of the NtD operator via potential theory, Nyström discretization, and the Cayley transform. At high frequencies (400 wavelengths across), with eigenfrequency relative error 10-10, we show that the method is 103 times faster than standard ones based upon a root search.
AB - We present a new algorithm for numerical computation of large eigenvalues and associated eigenfunctions of the Dirichlet Laplacian in a smooth, star-shaped domain in ℝd, d≥2. Conventional boundary-based methods require a root search in eigenfrequency k, hence take O(N3) effort per eigenpair found, where N=O(kd-1) is the number of unknowns required to discretize the boundary. Our method is O(N) faster, achieved by linearizing with respect to k the spectrum of a weighted interior Neumann-to-Dirichlet (NtD) operator for the Helmholtz equation. Approximations k^j to the square roots kj of all O(N) eigenvalues lying in [k-ε{lunate}, k], where ε{lunate}=O(1), are found with O(N3) effort. We prove an error estimate |k^j-kj|≤C(∈2k+∈3), with C independent of k. We present a higher-order variant with eigenvalue error scaling empirically as O(ε{lunate}5) and eigenfunction error as O(ε{lunate}3), the former improving upon the "scaling method" of Vergini and Saraceno. For planar domains (d=2), with an assumption of absence of spectral concentration, we also prove rigorous error bounds that are close to those numerically observed. For d=2 we compute robustly the spectrum of the NtD operator via potential theory, Nyström discretization, and the Cayley transform. At high frequencies (400 wavelengths across), with eigenfrequency relative error 10-10, we show that the method is 103 times faster than standard ones based upon a root search.
UR - http://www.scopus.com/inward/record.url?scp=84892372577&partnerID=8YFLogxK
U2 - 10.1002/cpa.21458
DO - 10.1002/cpa.21458
M3 - Article
SN - 0010-3640
VL - 67
SP - 351
EP - 407
JO - Communications on Pure and Applied Mathematics
JF - Communications on Pure and Applied Mathematics
IS - 3
ER -