No abstract

This work presents a study of the determination of uncertainties in [UTC − UTC(k)] needed for publication in the Bureau Inernational des Poids et Mesures's (BIPM's) Circular T and the Key Comparison Database, as required by the Mutual Recognition Arrangement. In the first part of the paper, an analytical solution based on the law of the propagation of uncertainty is derived. In the second part, the solution is verified numerically using the software used by the BIPM for the generation of UTC.

In Poland, an accurate two-way optical fibre time transfer (TWOTT) baseline of 420 km has been developed with a standard uncertainty of 112 ps. Meanwhile, the BIPM has been engaged in a pilot project to reduce the uncertainty of UTC GNSS time transfer calibrations from 5 ns to approximately 1.5 ns, for which it has developed a standard calibration scheme, named METODE, associated with an integrated portable GNSS calibration station. The TWOTT and the METODE respectively represent the world's most advanced time transfer method and calibration facility.In July 2013, the BIPM calibrator and the TWOTT self-calibrated optical fibre transceivers were set up side by side to compare their results. This is the first time that an operational TWOTT has been used in accurate time transfer and the first time that a GNSS calibration method has been validated with a technique accurate to 200 ps. As the two systems are completely independent and the latter is one order of magnitude more accurate than the first, this comparison allowed validation of the METODE and its uncertainty. It also confirmed that any bias in the TWOTT is at most 1 ns.

We refine our estimate of the uncertainty in [UTC - UTC(k)] by taking into account the contribution of correlations between the links. Using a matrix formulation to facilitate computation, we re-analyze the link-based uncertainties and obtain the same solution as found previously. We then evaluate the site-based uncertainties and compare the results with the link-based uncertainties.

No abstract

In a first attempt to evaluate performance of the Global Navigation Satellite System (GLONASS) precise-code time transfer, one-site measurements are used to show that single-channel GLONASS P-code, combined with temperature-stabilized antennas, reduces the noise experienced by time-receiving equipment to a few hundred picoseconds over an averaging time of one day, thus allowing frequency comparison at a level of a few parts in 10 15 .

No abstract

We present results from two calibrations performed in 2003 between the Global Positioning System (GPS) commonview receivers located at the National Institute of Standards and Technology (NIST) and at the Bureau National de Métrologie − Systèmes de Référence Temps Espace (BNM-SYRTE) of the Observatoire de Paris (OP) in the context of calibrations performed over 20 years, 1983-2003. We also present several years of continuous comparisons between receivers located at each of the two labs: NIST and OP. These results show that the best GPS receivers in use have delay variations with peak differences of under 5 ns over a year. This contributes to defining the current practical limits of GPS common-view time transfer. Since GPS commonview time transfer is still used for the majority of links between laboratories contributing to International Atomic Time (TAI), the noise and uncertainties in common-view affect the short-term performance of TAI, for averaging times from 5 to 30 days.