Files in this item
|(no description provided)|
|Title:||Recursive algorithms for computational electromagnetics|
|Doctoral Committee Chair(s):||Chew, Weng Cho|
|Department / Program:||Electrical and Computer Engineering|
|Degree Granting Institution:||University of Illinois at Urbana-Champaign|
|Subject(s):||Engineering, Electronics and Electrical|
|Abstract:||Efficient and fast recursive algorithms for both the spectral-domain and the space-domain solutions of the electromagnetic scattering problems have been developed. These algorithms have less than $O(N\sp3)$ computational complexities and less than $O(N\sp2)$ memory requirements for arbitrary geometries of scatterers clustered together. Although the algorithms are discussed as they are applied to the electromagnetic scattering problems, their domains of applicability can be extended to other types of electromagnetic problems (e.g., guidance, resonance, and radiation) and also to other types of field and wave equations (e.g., acoustic, elastic, and Schrodinger).
The applications of these algorithms to the conducting strip and patch geometries have been demonstrated. Dielectric and magnetic materials can also be incorporated into conductor geometries. Due to the availability of the spectral Green's function, spectral-domain algorithms can efficiently handle geometries consisting of an arbitrary number of infinitely thin, conducting strips and flat patches of any shape embedded in arbitrarily layered media, in which the layers are infinitely large in the transverse directions. On the other hand, the space-domain algorithms presented in this dissertation are the recursive T-matrix algorithms, and due to the nature of the T-matrix formulations, they can easily handle geometries consisting of conductors, dielectrics and magnetic materials of finite size. Therefore, various problems defined in broad classes of geometries can be solved with these spectral-domain and space-domain algorithms.
A recursive implementation of the method of moments is also presented. This algorithm is based on the principle of inversion of a general matrix by partitioning. Since it is a recursive algorithm, and since each recursion step requires $O(N\sp2)$ operations, it can efficiently solve the problems in which one has to modify or perturb some parts of a main body whose solution is already known.
When solving the electromagnetic scattering problem, these algorithms give the full-wave solution without having to make any approximations. Being computational algorithms, they are "exact" in the numerical sense. These algorithms also give the solution for all possible incident waves or "right-hand sides" at once, a property that is not shared by some other fast solution techniques such as the conjugate-gradient method. Furthermore, as opposed to some other formulation schemes such as the finite-element method, these algorithms naturally incorporate the radiation condition at infinity; therefore, they can handle geometries in unbounded media.
|Rights Information:||Copyright 1991 Gurel, Levent|
|Date Available in IDEALS:||2011-05-07|
|Identifier in Online Catalog:||AAI9210822|
This item appears in the following Collection(s)
Graduate Dissertations and Theses at Illinois
Graduate Theses and Dissertations at Illinois
Dissertations and Theses - Electrical and Computer Engineering
Dissertations and Theses in Electrical and Computer Engineering