Efficient single-step preconditioned HSS iteration methods for complex symmetric linear systems

摘要

We propose a single-step preconditioned variant of HSS (SPHSS) and an efficient parameterized SPHSS (PSPHSS) iteration method for solving a class of complex symmetric linear systems. Under suitable conditions, we analyze the convergence properties of the SPHSS and PSPHSS iteration methods. Theoretical analysis shows that the minimal upper bounds for the spectral radius of the SPHSS and PSPHSS iteration matrices are less than those of the SHSS and PSHSS iteration matrices when using the optimal parameters, respectively. Numerical results show that the PSPHSS iteration method has comparable advantage over several other iteration methods whether the experimental optimal parameters are used or not. (C) 2017 Elsevier Ltd. All rights reserved.