A Spectral Multiresolution Preconditioner Combined with Sparse-Matrix/Canonical Grid Method for Analysis of Microstrip Circuits

摘要

A spectral multiresolution (MR) preconditioner combined with Sparse-Matrix/Canonical Grid Method (SMCG) is developed for fast analysis of microstrip circuits structures. The SMCG is used to speed up the matrix-vector product operations. The spectral MR preconditioner combined with a diagonal preconditioning is employed to achieve the fast convergence rate by augmenting to the generalized minimal residual iterative method (GMRES) a few eigenvectors associated with the smallest eigenvalues of the system. Numerical experiments demonstrate the accuracy and efficiency of the proposed method.