NOISE FILTERING OF SEISMIT DATA

Adaptive filtering method to remove ground roll from seismic data. In an M channel adaptive filter, weights Wi are set using an adaptive algorithm based on seeking the minimum in the partial differential of cost function J. The cost function includes an expansion of the primary trace d into d=dg+Δd (where: dg is ground roll contribution and Δd=dsig+dran, where dsig is the reflected signal component and dran is a random noise component) and a corresponding expansion of the reference x into x=xg+Δx (where xg is ground roll contribution and Δx=xsig+xran; where xsig is a reflected signal component and xran is a random noise component). The delta components are included in the denominator of cost function J to provide an optimal solution of the filter coefficients biased by the reflection signal and random noise is removed.