DocumentCode
822022
Title
Efficient modeling of microwave integrated-circuit geometries via a dynamically adaptive mesh Refinement-FDTD technique
Author
Liu, Yaxun ; Sarris, Costas D.
Author_Institution
Edward S. Rogers Sr. Dept. of Electr. & Comput. Eng., Univ. of Toronto, Ont., Canada
Volume
54
Issue
2
fYear
2006
Firstpage
689
Lastpage
703
Abstract
The finite-difference time-domain (FDTD) method is combined with an adaptive mesh refinement (AMR) technique, to achieve a fast, time-domain solver for Maxwell´s equations (AMR-FDTD), based on a three-dimensional moving/rotating Cartesian mesh. This combination allows the proposed solver to adapt to the problem at hand, optimally distributing computational resources in a given domain as needed, by recursively refining a coarse grid in regions of large over time gradient of electromagnetic field energy. Several applications of the method to the analysis of microwave circuit geometries demonstrate its salient features and its outstanding efficiency as a microwave computer-aided design tool.
Keywords
Maxwell equations; circuit CAD; finite difference time-domain analysis; gradient methods; integrated circuit design; integrated circuit modelling; mesh generation; microwave integrated circuits; 3D Cartesian mesh; Maxwell equations; adaptive mesh refinement; coarse grid definition; electromagnetic field energy gradient; finite-difference time-domain method; microwave circuit geometries; microwave computer-aided design tool; microwave integrated circuit modeling; time-domain solver; Adaptive mesh refinement; Distributed computing; Electromagnetic fields; Finite difference methods; Geometry; Grid computing; Maxwell equations; Microwave theory and techniques; Solid modeling; Time domain analysis; Adaptive mesh refinement (AMR); finite difference time domain (FDTD); microwave integrated circuits;
fLanguage
English
Journal_Title
Microwave Theory and Techniques, IEEE Transactions on
Publisher
ieee
ISSN
0018-9480
Type
jour
DOI
10.1109/TMTT.2005.862660
Filename
1589493
Link To Document