Error Estimates for Regularization Methods in Hilbert Scales
Error Estimates for Regularization Methods in Hilbert Scales
复制标题
希尔伯特量表中正则化方法的误差估计
DOI:
10.1137/s0036142994269411
复制
发表时间:
1996
影响因子:
2.9
通讯作者:
U. Tautenhahn
中科院分区:
文献类型:
--
作者:
U. Tautenhahn
In this paper we study regularization methods to reconstruct the solution $x^*$ of the linear ill-posed problem $Ax=y$, $A: X \to Y$, from noisy data $y^\delta \in Y$. The regularization methods are of the general form $x_\alpha^\delta = \bar{x} + g_\alpha (B^{-s}A^*A)B^{-s}A^* (y^\delta -A\bar{x})$ where $B$ denotes an unbounded self-adjoint strictly positive definite operator in the Hilbert space $X$. Assuming $\|Ax\| \sim \|x\|_{-a}$ and $\|x^*-\bar{x}\|_p \le E$ for some $\bar{x} \in X$, $a \ge 0$ and $p \ge 0$ ($\|x\|_r = \|B^{r/2} x\|$ is the norm in a Hilbert scale $(X_r)_{r \in \rnus}$) we derive error estimates which show that the accuracy of the above regularization methods is order optimal provided that the function $g_\alpha (\lambda)$, the regularization parameter $\alpha$, and the parameter $s$ are chosen properly.