Signal Restoration Combining Modified Tikhonov Regularization and Preconditioning Technology

被引：6
|
作者：
Dou, Hong-Xia [1 ]
Li, Hou-Biao [1 ]
Fans, Qing-Yu [1 ]
Chen, Yi-Cheng [2 ]
机构：
[1] Univ Elect Sci & Technol China, Sch Math Sci, Chengdu 611731, Sichuan, Peoples R China
[2] Lehigh Univ, Dept Elect & Comp Engn, Bethlehem, PA 18015 USA
来源：
IEEE ACCESS | 2017年 / 5卷
关键词：
Signal deconvolution; modified Tikhonov regularization; preconditioner; KRONECKER PRODUCT APPROXIMATIONS; ILL-POSED PROBLEMS; IMAGE-RESTORATION; BOUNDARY-CONDITIONS; TOEPLITZ MATRICES; L-CURVE;
D O I：
10.1109/ACCESS.2017.2767702
中图分类号：
TP [自动化技术、计算机技术];
学科分类号：
0812 ;
摘要：
The purpose of signal restoration is to acquire a clean signal from the degraded signal which contains blur and noise. In this paper, a modified Tikhonov regularization method based on the standard Tikhonov regularization matrix is proposed, and the corresponding preconditioner is designed to accelerate the convergence of the proposed algorithm. The proposed method shows the best performance than several competitive methods. In addition, the convergence speed is improved significantly.
引用
收藏
页码：24275 / 24283
页数：9
相关论文
共 50 条
12345