Return to search

Cauchy interpolation for multi-variate and multi-derivative data

There is often a need to interpolate data that is obtained through experiment or computational analysis, because the data is difficult or expensive to obtain. An example is the scattering parameters of microwave devices, obtained through computationally intensive finite element (FE) analysis. Cauchy interpolation is an established solution to this problem. In this thesis it is extended to interpolate data over a multi-parameter space, when the data available includes not just the function to be interpolated, but also its derivatives with respect to each parameter. The finite element method (FEM) provides such derivatives. The new algorithm is applied to a simple RLC circuit test case, and to real data from a 3D FE analysis of a rectangular waveguide component, in a 4-parameter space. Results show the effectiveness of the approach taken.

Identiferoai:union.ndltd.org:LACETR/oai:collectionscanada.gc.ca:QMM.100227
Date January 2007
CreatorsKaufman, Jonathan, 1981-
PublisherMcGill University
Source SetsLibrary and Archives Canada ETDs Repository / Centre d'archives des thèses électroniques de Bibliothèque et Archives Canada
LanguageEnglish
Detected LanguageEnglish
TypeElectronic Thesis or Dissertation
Formatapplication/pdf
CoverageMaster of Engineering (Department of Electrical and Computer Engineering.)
Rights© Jonathan Kaufman, 2007
Relationalephsysno: 002668635, proquestno: AAIMR38484, Theses scanned by UMI/ProQuest.

Page generated in 0.0029 seconds