Files in this item



application/pdf3101856.pdf (6MB)Restricted to U of Illinois
(no description provided)PDF


Title:A Study of MLFMA for Large -Scale Scattering Problems
Author(s):Hastriter, Michael Larkin
Doctoral Committee Chair(s):Chew, Weng Cho
Department / Program:Electrical Engineering
Discipline:Electrical Engineering
Degree Granting Institution:University of Illinois at Urbana-Champaign
Subject(s):Physics, Electricity and Magnetism
Abstract:This research is centered in computational electromagnetics with a focus on solving large-scale problems accurately in a timely fashion using first principle physics. Error control of the translation operator in 3-D is shown. A parallel implementation of the multilevel fast multipole algorithm (MLFMA) was studied as far as parallel efficiency and scaling. The large-scale scattering program (LSSP), based on the ScaleME library, was used to solve ultra-large-scale problems including a 200lambda sphere with 20 million unknowns. As these large-scale problems were solved, techniques were developed to accurately estimate the memory requirements. Careful memory management is needed in order to solve these massive problems. The study of MLFMA in large-scale problems revealed significant errors that stemmed from inconsistencies in constants used by different parts of the algorithm. These were fixed to produce the most accurate data possible for large-scale surface scattering problems. Data was calculated on a missile-like target using both high frequency methods and MLFMA. This data was compared and analyzed to determine possible strategies to increase data acquisition speed and accuracy through multiple computation method hybridization.
Issue Date:2003
Description:137 p.
Thesis (Ph.D.)--University of Illinois at Urbana-Champaign, 2003.
Other Identifier(s):(MiAaPQ)AAI3101856
Date Available in IDEALS:2015-09-25
Date Deposited:2003

This item appears in the following Collection(s)

Item Statistics