English

Multidimensional Phase Recovery and Interpolative Decomposition Butterfly Factorization

Numerical Analysis 2020-04-22 v2 Numerical Analysis

Abstract

This paper focuses on the fast evaluation of the matvec g=Kfg=Kf for KCN×NK\in \mathbb{C}^{N\times N}, which is the discretization of a multidimensional oscillatory integral transform g(x)=K(x,ξ)f(ξ)dξg(x) = \int K(x,\xi) f(\xi)d\xi with a kernel function K(x,ξ)=e2π\iΦ(x,ξ)K(x,\xi)=e^{2\pi\i \Phi(x,\xi)}, where Φ(x,ξ)\Phi(x,\xi) is a piecewise smooth phase function with xx and ξ\xi in Rd\mathbb{R}^d for d=2d=2 or 33. A new framework is introduced to compute KfKf with O(NlogN)O(N\log N) time and memory complexity in the case that only indirect access to the phase function Φ\Phi is available. This framework consists of two main steps: 1) an O(NlogN)O(N\log N) algorithm for recovering the multidimensional phase function Φ\Phi from indirect access is proposed; 2) a multidimensional interpolative decomposition butterfly factorization (MIDBF) is designed to evaluate the matvec KfKf with an O(NlogN)O(N\log N) complexity once Φ\Phi is available. Numerical results are provided to demonstrate the effectiveness of the proposed framework.

Keywords

Cite

@article{arxiv.1908.09376,
  title  = {Multidimensional Phase Recovery and Interpolative Decomposition Butterfly Factorization},
  author = {Ze Chen and Juan Zhang and Kenneth L. Ho and Haizhao Yang},
  journal= {arXiv preprint arXiv:1908.09376},
  year   = {2020}
}
R2 v1 2026-06-23T10:56:18.781Z