2017 Volume 14 Issue 8 Pages 20160791
It is well known that the impedance matrix equation obtained by discretizing the electric field integral equation (EFIE) is dense and ill-conditioned. In this paper, a combination of the Sherman-Morrison-Woodbury (SMW) formula-based preconditioner and the adaptive cross approximation (ACA) algorithm is proposed to accelerate the iterative solution of the EFIE. The SMW preconditioner (SMWP) and the ACA algorithm are employed to reduce the number of iterations and the complexity of a matrix-vector product (MVP), respectively. The SMWP can efficiently construct a sparse approximate inverse of the impedance matrix with the aid of the SMW formula. It can be considered as an improved version of the conventional block diagonal preconditioner (BDP) and has better performance. Numerical results are provided to demonstrate that the hybrid SMWP-ACA is effective to reduce the iterative solution time.
