Least-squares migration/inversion of blended data

direct imaging through inversion

A way to reduce the crosstalk is to go beyond migration by formulating the imaging problem as a LSI problem. The motivation behind LSI is that the pseudo-inverse of the combined modeling operator $\widetilde{\bf L}$ should be able to optimally reconstruct the reflectivity; hence the image would be minimally affected by the crosstalk artifacts. The LSI can be performed either in the model space or in the data space, each of which has its own pros and cons. We will analyze both of them for imaging the blended data.