# Translations:Wavelet processing/9/en

In summary, wavelet processing involves filtering the seismic trace to change the shape of the underlying seismic wavelet in advantageous ways. This filtering operation is implemented with a *shaping filter*. The physical field wavelet is a one-sided (or causal) signal whose initial nonzero value occurs at the arrival time of the reflected event. Because the earth necessarily acts as a causal filter, clearly it is impossible for the physical wavelet to have a precursor (i.e., nonzero amplitudes) arriving *before* the wavelet’s actual arrival time. However, the seismic wavelet takes some time (a few milliseconds or so) to build up to its peak amplitude, and thus this peak amplitude occurs after the wavelet’s arrival time. Almost always, some secondary peaks follow the main wavelet peak. The objective of wavelet processing is to convert the physical field seismic wavelet into an artificial interpreter wavelet, which is more easily pickable and hence more pleasing to the eye of an interpreter. Like great art, this artifact does not have to be a wavelet that satisfies physical laws, but it must be esthetically pleasing.