next up previous print clean
Next: Acknowledgments Up: Guitton and Verschuur: adaptive Previous: Prestack land data multiple


Significant amplitude differences between signal and noise make the $\ell^2$-norm an unsuitable choice to estimate shaping filters. We showed that the $\ell^1$-norm should always be considered in these circumstances. In addition, applications to real data with surface-related multiples tend to prove that the $\ell^1$-norm should also be considered in the simplest cases.

Stanford Exploration Project