Fast Iterative Solution of Sparsely Sampled Seismic Inverse Problems

摘要

In linearized seismic inverse scattering, one of the problems that has received little attention so far is the existence of large gaps in the acquisition geometry due to the use of a limited number of sources and receivers. Frequently used Born inversion methods, based on high-frequency asymptotics, do not take these kind of sampling effects into account. Therefore, especially for three-dimensional problems, the results may suffer from serious artifacts. Inverse scattering methods, based on the iterative minimization of some error norm, are more accurate for sparsely sampled measures but require a prohibitive amount of computational effort unless the rate of convergence is extremely fast. In the present paper, it is shown how the convergence of iterative scattering methods is accelerated significantly by using the above mentioned (asymptotic) Born inversion as a preconditioner. (Copyright (c) 1993 by Faculty of Technical Mathematics and Informatics, Delft, The Netherlands.)