OBC signal fidelity

A suite of three tests was performed to characterize the signal fidelity of OBC 4C acquisition systems. The test methodology was to evaluate individual sensor stations by acquiring source lines that were parallel to the in‐line and cross‐line horizontal sensors and source lines that were at 45° to the in‐line and cross‐line sensors. This technique provides constant azimuth gathers with a uniform offset range and removes issues associated with source array directivity.

Characterization of the test data identified the frequency content of the geophone signals and the correlation between the vertical and cross‐line geophones as the most sensitive indicators of signal infidelity. In the former case, stations with questionable signal fidelity exhibited a very reverberatory signal. This signal was most evident on the cross‐line sensor. In the latter case, when normalized cross‐correlation coefficients are computed in a moving window, the cross‐line sensor and the vertical sensor are highly correlated, beginning several hundred milliseconds after the first arrivals.

These characteristics can be exploited to allow stations with questionable signal fidelity to be programmatically identified. One means of identifying questionable stations is to compute the histogram of the instantaneous frequency. The frequency distributions from questionable stations are unambiguously distinguishable from stations that exhibit better signal fidelity. It was noted that signal fidelity appeared as a range, between acceptable and poor. To characterize the signal fidelity of an acquisition system adequately, the number of test samples must be statistically significant.