Preconditioning Toeplitz-plus-diagonal linear systems using the Sherman-Morrison-Woodbury formula

详细信息    查看全文

• 作者：Chaojie Wang ^{wangcjbuaa@163.com" class="auth_mail" title="E-mail the corresponding author} ; Hongyi Li ; Di Zhao ; ^{zdbuaa@163.com" class="auth_mail" title="E-mail the corresponding author}
• 关键词：Toeplitz-plus-diagonal linear systems ; Approximate inverse preconditioner ; Incomplete factorization ; Sherman&ndash ; Morrison&ndash ; Woodbury formula
• 刊名：Journal of Computational and Applied Mathematics
• 出版年：2017
• 出版时间：1 January 2017
• 年：2017
• 卷：309
• 期：Complete
• 页码：312-319
• 全文大小：991 K

文摘

In order to solve the Toeplitz-plus-diagonal linear systems arising from image restorations efficiently, we propose a sparse approximate inverse preconditioner based on the Sherman–Morrison–Woodbury formula. The preconditioner can be constructed through an incomplete factorization combined with some dropping strategies. When the preconditioner is applied to the conjugate gradient method for solving the Toeplitz-plus-diagonal linear systems, numerical results show that our preconditioning method is more effective than other existing ones.