Full text loading...
-
Predictive Deconvolution Revisited with Neural Nets
- Publisher: European Association of Geoscientists & Engineers
- Source: Conference Proceedings, 57th EAGE Conference and Exhibition, May 1995, cp-90-00296
- ISBN: 978-90-73781-06-1
Abstract
Recent work has shown the power of neural nets to solve problems in signal processing. Traditional predictive deconvolution filter design rests on the assumption that the seismic trace is a sample from a weakly stationary process, and that the sequence of subsurface reflection coefficients is random and uncorrelated.