Study of statistical approaches to the solution of linear discrete and integral problems

Author

Maffezzoni, Paolo ; Brambilla, Angelo

Author_Institution

Dipt. di Elettronica e Inf., Politecnico di Milano, Italy

Volume

50

Issue

9

fYear

2003

Firstpage

1153

Lastpage

1161

Abstract

A study of convergence and accuracy properties of the fixed random walk (RW) and the floating RW algorithms is presented. These two algorithms can be considered as a statistical reinterpretation of deterministic processes and in particular of the Jacobi´s iterative method. Reinterpretation allows a better insight into the two statistical algorithms and mainly the derivation of useful expressions for the errors introduced by them.

Keywords

convergence of numerical methods; deterministic algorithms; error analysis; iterative methods; linear algebra; random processes; Jacobi iterative method; convergence; deterministic process; error analysis; fixed random walk; floating random walk; linear discrete problem; linear integral problem; statistical algorithm; Computational geometry; Computational modeling; Electrostatics; Integral equations; Iterative algorithms; Iterative methods; Jacobian matrices; Laplace equations; Parasitic capacitance; Ultra large scale integration;

fLanguage

English

Journal_Title

Circuits and Systems I: Fundamental Theory and Applications, IEEE Transactions on

Publisher

ieee

ISSN

1057-7122

Type

jour

DOI

10.1109/TCSI.2003.816304

Filename

1230223