Regularization parameter estimation for large-scale Tikhonov regularization using a priori information

Solutions of numerically ill-posed least squares problems for by Tikhonov regularization are considered. For , the Tikhonov regularized least squares functional is given by where matrix W is a weighting matrix and is given. Given a priori estimates on the covariance structure of errors in the measurement data , the weighting matrix may be taken as which is the inverse covariance matrix of the mean 0 normally distributed measurement errors in . If in addition is an estimate of the mean value of , and [sigma] is a suitable statistically-chosen value, J evaluated at its minimizer approximately follows a [chi]2 distribution with degrees of freedom. Using the generalized singular value decomposition of the matrix pair , [sigma] can then be found such that the resulting J follows this [chi]2 distribution. But the use of an algorithm which explicitly relies on the direct solution of the problem obtained using the generalized singular value decomposition is not practical for large-scale problems. Instead an approach using the Golub-Kahan iterative bidiagonalization of the regularized problem is presented. The original algorithm is extended for cases in which is not available, but instead a set of measurement data provides an estimate of the mean value of . The sensitivity of the Newton algorithm to the number of steps used in the Golub-Kahan iterative bidiagonalization, and the relation between the size of the projected subproblem and [sigma] are discussed. Experiments presented contrast the efficiency and robustness with other standard methods for finding the regularization parameter for a set of test problems and for the restoration of a relatively large real seismic signal. An application for image deblurring also validates the approach for large-scale problems. It is concluded that the presented approach is robust for both small and large-scale discretely ill-posed least squares problems.