An adaptive moving window multiple sidelobe canceller for seismic data
G. Robertson, Rodney Lynn Kirlin, W.-S. Lu · 2002
The multiple sidelobe canceller (MSG) has a higher resolution than the more common semblance algorithm allowing it to reveal more details in the data (i.e. seismic data). In a conventional MSC the calculation of a pseudoinverse of a matrix makes the algorithm very slow for repeated applications. A method is presented to reduce this computational complexity of the MSC when applied to a moving window. Two updating procedures using the change in the data from one window position to the next are shown to reduce the computational complexity of the MSC by more than 75% for a downward movement (with time) and more than 55% for a sideways movement (across traces) of the window. In addition these methods for updating the pseudoinverse appear to produce more accurate results.