Comp. Appl. Math.
DOI 10.1007/s40314-015-0231-6
A non-alternating preconditioned HSS iteration method
for non-Hermitian positive definite linear systems
Yu-Jiang Wu
1,2,3
· Xu Li
4,5
· Jin-Yun Yuan
6
Received: 30 June 2014 / Revised: 9 April 2015 / Accepted: 10 April 2015
© SBMAC - Sociedade Brasileira de Matemática Aplicada e Computacional 2015
Abstract By utilizing the preconditioned Hermitian and skew-Hermitian splitting (PHSS)
iteration technique, we establish a non-alternating PHSS (NPHSS) iteration method for solv-
ing large sparse non-Hermitian positive definite linear systems. The convergence analysis
demonstrates that the iterative series produced by the NPHSS method converge to the unique
solution of the linear system when the parameters satisfy some moderate conditions. We also
give a possible optimal upper bound for the iterative spectral radius. Moreover, to reduce the
computational cost, we establish an inexact variant of the NPHSS (INPHSS) iteration method
whose convergence property is studied. Both theoretical and numerical results validate that
the NPHSS method outperforms the PHSS method when the Hermitian part of the coefficient
matrix is dominant.
Communicated by Ya-xiang Yuan.
B Xu Li
mathlixu@163.com; lixu@lut.cn
Yu-Jiang Wu
myjaw@lzu.edu.cn
Jin-Yun Yuan
jin@ufpr.br
1
School of Mathematics and Statistics, Lanzhou University, Lanzhou 730000, China
2
Gansu Key Laboratory of Applied Mathematics and Complex Systems,
Lanzhou 730000, China
3
Institute of Industrial Mathematics, IMPA, CP: 19.081, Curitiba, PR 81531-980, Brazil
4
College of Electrical and Information Engineering, Lanzhou University of Technology,
Lanzhou 730050, China
5
Department of Applied Mathematics, Lanzhou University of Technology, Lanzhou 730050, China
6
Department of Mathematics, Centro Politécnico, Federal University of Paraná, CP: 19.081, Curitiba,
PR 81531-980, Brazil
123