Noise characteristics of high-frequency, short-duration GPS records from analysis of identical, collocated instruments

High-rate (>1 Hz) GPS, currently used for measurement of dynamic processes and as a universal time and frequency standard, is usually assumed to be affected by site-specific errors and only limited studies of the error properties of high-frequency (10 Hz), short-duration (<100 s) GPS records exist. Preliminary studies, provided evidence of instrument-specific errors, which were investigated on the basis of systematic experiments with various types of stationary, collocated, identical GPS units. This analysis revealed differences between couples of receiver/antenna units, while spectral analyses revealed low-frequency colored noise, statistically significant below 0.2 Hz and, gradually decaying with increasing duration of observations, so that above 2.5 Hz measurements are contaminated only by white noise.These results explain why long duration GPS records and the high-frequency component of very short-duration records are practically free of colored noise and why spectra of noisy records are accurate, and hence indicate that GPS is suitable for monitoring of dynamic effects.

► Comparison of high rate GPS measurements from identical, collocated receivers.
► Differences between measurements found in raw and processed data.
► Differences due to colored noise affecting low frequencies.
► High frequencies and long-duration records affected only by white noise.
► High-frequency event measurements or long duration records free of colored noise.