On the role of diffractions in velocity model building: a full-waveform inversion example

Studia Geophysica et Geodaetica - Imaging of small-scale heterogeneities is important for the geological exploration in complex environments. It requires a processing sequence tuned to...     

Co-Phased Median Filters, Some Peculiarities of Sweep Signal Processing

Median filters are nonlinear and a theoretical analysis of their behavior is very difficult and so are rarely used for the processing of seismic data. However, they are able to preserve steps, sharp discontinuities and edges that are lost using most other standard filters. As seismic data are mostly harmonic or frequency modulated signals in the frequency range 1 to 10 Hz, the median filter must be adapted so as to suppress noise but not unduly distort the signal. This can be accomplished by using a median filter whose length is n times that of signal period where n is even. By use of weighted-order statistics, samples of the signals closest in phase can be obtained. To illustrate the method, a signal from a frequency band of 6.3–7.5 Hz was processed and the quality of the signal enhanced two fold over the quality of the signal without processing.