ICA Separation for Improving Multiple Subtraction

The suppression of multiple events is a crucial task in seismic data processing, and the removal of the predicted multiples from real data is recognized as one of the main challenges for the success of the SRME technique. The traditional least-squares matching approach can affect the primary events, since the estimated multiples tend to adapt to the primaries under the minimum energy condition. Therefore, we propose to improve the multiple removal results by replacing the multiple subtraction by a separation step using Independent Component Analysis (ICA) methods. We apply the ICA method as a post-processing technique, to be used after adaptive filtering: primaries and multiples can be separated by computing the optimal rotation between these two components, after adaptive filtering, by exploiting the non-Gaussianity of the signals involved. Moreover, we apply the ICA method in local 2D time-space windows to better compensate the space and time variant character of the data. The method is tested on a real marine dataset after 2D LS adaptive subtraction producing promising results.