A Refinement of some Overrelaxation Algorithms for Solving a System of Linear Equations

Kyurkchiev, Nikolay, and Iliev, Anton. "A Refinement of some Overrelaxation Algorithms for Solving a System of Linear Equations." Serdica Journal of Computing 7.3 (2013): 245-256. <http://eudml.org/doc/268658>.

@article{Kyurkchiev2013,
abstract = {In this paper we propose a refinement of some successive overrelaxation methods based on the reverse Gauss–Seidel method for solving a system of linear equations Ax = b by the decomposition A = Tm − Em − Fm, where Tm is a banded matrix of bandwidth 2m + 1. We study the convergence of the methods and give software implementation of algorithms in Mathematica package with numerical examples. ACM Computing Classification System (1998): G.1.3.This paper is partly supported by project NI13 FMI–002 of Department for Scientific Research, Paisii Hilendarski University of Plovdiv.},
author = {Kyurkchiev, Nikolay, Iliev, Anton},
journal = {Serdica Journal of Computing},
keywords = {reverse Gauss–Seidel method; Nekrassov–Mehmke 2 method – (NM2); Successive Overrelaxation method with 1 parameter; based on (NM2) – (SOR1NM2); Successive Overrelaxation method with 2 parameters; based on (NM2) – (SOR2NM2); Refinement of (SOR1NM2)},
language = {eng},
number = {3},
pages = {245-256},
publisher = {Institute of Mathematics and Informatics Bulgarian Academy of Sciences},
title = {A Refinement of some Overrelaxation Algorithms for Solving a System of Linear Equations},
url = {http://eudml.org/doc/268658},
volume = {7},
year = {2013},
}

TY - JOUR
AU - Kyurkchiev, Nikolay
AU - Iliev, Anton
TI - A Refinement of some Overrelaxation Algorithms for Solving a System of Linear Equations
JO - Serdica Journal of Computing
PY - 2013
PB - Institute of Mathematics and Informatics Bulgarian Academy of Sciences
VL - 7
IS - 3
SP - 245
EP - 256
AB - In this paper we propose a refinement of some successive overrelaxation methods based on the reverse Gauss–Seidel method for solving a system of linear equations Ax = b by the decomposition A = Tm − Em − Fm, where Tm is a banded matrix of bandwidth 2m + 1. We study the convergence of the methods and give software implementation of algorithms in Mathematica package with numerical examples. ACM Computing Classification System (1998): G.1.3.This paper is partly supported by project NI13 FMI–002 of Department for Scientific Research, Paisii Hilendarski University of Plovdiv.
LA - eng
KW - reverse Gauss–Seidel method; Nekrassov–Mehmke 2 method – (NM2); Successive Overrelaxation method with 1 parameter; based on (NM2) – (SOR1NM2); Successive Overrelaxation method with 2 parameters; based on (NM2) – (SOR2NM2); Refinement of (SOR1NM2)
UR - http://eudml.org/doc/268658
ER -