Next: Marine Data Results
Up: THEORY
Previous: Hybrid l-l function
My goal in this paper is to compare three different inversion schemes for the multiples attenuation
problem. They all aim to produce a velocity model where primaries are muted out
and the predicted multiples are subtracted from the original data. I successively solve
- 1.
,- 2.
, - 3.
,
and compare the results. I call arbitrarily ``l1 norm'' any Huber function with
a small threshold. Let us assume now that to the l1 norm, for the data residual, corresponds
a threshold
![\begin{displaymath}
\epsilon = \frac{max\vert\bold{d}\vert}{100}.\end{displaymath}](img15.gif)
In addition, for the regularization term, let us say that to the l1 norm corresponds a threshold
![\begin{displaymath}
\epsilon = \frac{max\vert\bold{d}\vert}{10000}.\end{displaymath}](img16.gif)
is chosen smaller than before leading to a larger l1 treatment of the model.
I show later on that the convergence is greatly reduced by the addition of this regularization term.
Next: Marine Data Results
Up: THEORY
Previous: Hybrid l-l function
Stanford Exploration Project
4/27/2000