A hybrid preconditioner of banded matrix approximation and alternating direction implicit iteration for symmetric Sinc--Galerkin linear systems (Q1874674)

The authors investigate the solution of a system of linear equations: \( (\mathbf{\Psi}_x \otimes D_y+D_x\otimes\mathbf{\Psi}_y)\mathbf{u}=\mathbf{g}, \) where \(\otimes\) is the Kronecker product symbol, \(\mathbf{\Psi}_x\) and \(\mathbf{\Psi}_y\) are Toeplitz-plus-diagonal matrices, and \(D_x\) and \(D_y\) are diagonal matrices. The system arises from symmetric Sinc-Galerkin discretization of a separable second selfadjoint elliptic boundary value problem. The main aim of this paper is to study the preconditioning properties of the two-step preconditioner based on the banded matrix approximation and the alternating direction implicit iteration for these Sinc-Galerkin systems. Numerical examples show that the new preconditioner is practical and efficient to precondition the conjugate gradient method for solving the above symmetric Sinc-Galerkin linear system. Related results can be found in the papers of \textit{J. Lund} [Math. Comput. 47, 571-588 (1986; Zbl 0629.65085)] and of \textit{M. Ng} [IMA J. Numer. Anal. 19, 357-373 (1999; Zbl 0952.65057)].