Iteratively Separating the Simultaneous Seismic Data
Y. Chen, S. Fomel and K. Schleicher
Event name: 1st Latin American Geosciences Student Conference (LAGSC)
Session: Latin American Geosciences Student Conference
Publication date: 08 April 2013
Info: Extended abstract, PDF ( 1.33Mb )
Price: € 20
Simultaneous source technique has been attracting more and more attention becasue of its intuitive benefit in largely improving acquistion efficiency and somewhat resulting in higher record quality. However, the intense crosstalk between two or more sources lays a challenge for the processing afterwards, which calls for more specific and robust processing methods. In this paper, we introduce a new iteration based estimation scheme for the separation of the blended seismic data. We first construct a common estimation problem, then we use shaping regularization to constrain the model when iteratively solving the problem. In our method, we use soft thresholding in the seislet domain as our shaper to remove crosstalk noise and at the same time preserve the useful dip components in the seismic data. We use a numerically blended complex synthetic example to demostrate the performance of the proposed algorithm.