Effects of Data Processing and Environmental Conditions on the Accuracy of the USNO Timescale

摘要

Humidity has significant long-term effects on cesium clock rates that may be controlled environmentally, but not during data processing. Allan variances are minimized at a temperature depending on the clock and clock type. There is no dependence of Allan variance on manufacturing batch. A surprisingly large fraction of clock rate and variance changes may be attributed to human interference or the need for it. Little or no improvement is obtained by altering the unity-or-zero weighting scheme of the current USNO timescale algorithm. An algorithm based on robust ARIMA modelling yields a timescale that may differ markedly and is in most respects inferior to that generated by the current algorithm. The NIST algorithm is comparable in accuracy and stability to the current algorithm, except on the short term, where it is significantly less stable.