## Abstract A single integral equation (SIE) method is developed to analyze the electromagnetic scattering from 3D composite conducting and dielectric bodies. Compared with the conventional coupled integral equation (CIE) method, the SIE method reduces the number of unknowns by a factor of two for
Integration and interpolation based on fast spherical transforms for the multilevel fast multipole method
β Scribed by Indranil Chowdhury; Vikram Jandhyala
- Publisher
- John Wiley and Sons
- Year
- 2006
- Tongue
- English
- Weight
- 111 KB
- Volume
- 48
- Category
- Article
- ISSN
- 0895-2477
No coin nor oath required. For personal study only.
β¦ Synopsis
This study presents the application of fast spherical transforms developed by Driscoll and Healy (Adv Appl Math 15 (1994), 202-250) to the full-wave multilevel fast multipole method. An accurate uniformgrid based quadrature rule is presented, along with fast algorithms for interpolation and anterpolation. Error behavior and algorithmic complexities are discussed. Finally an efficient method for the generation of translation operators is presented. The overall method results in efficient and error-controllable schemes for quadrature and interpolation.
π SIMILAR VOLUMES
## Abstract This article presents an efficient hybrid method with the IEβFFT algorithm for solving scattering and radiation problems of wire antennas mounted on electrically large platforms. To model arbitrary metallic structures, including wire antennas and both open and closed surfaces, in this m
A noΒ¨el method combining the conjugate-gradient and ( ) nonuniform fast Fourier transform CGαNUFFT algorithms for solΒ¨ing integral equations in electromagnetics is presented. The CGαNUFFT method has been applied to one-and two-dimensional inhomogeneous media. Numerical results show the adΒ¨antages of
The nearest neighbor rule or k-nearest neighbor rule is a technique of nonparametric pattern recognition. Its algorithm is simple and the error is smaller than twice the Bayes error if there are enough training samples. However, it requires an enormous amount of computation, proportional to the numb