Carsten Rieck, RISE, Kenneth Jaldehag, RISE, Sweden

View Abstract Sign in for premium content


GNSS remains to be one of the most used sources of distributed time and continues to be the most widely used instrument for metrological time comparisons. Requirements on calibrations are increasing, partly due to new emerging application requirements, and partly due to new requirements on legal traceability to UTC. NMIs are providing a direct access to UTC via their UTC(k) realizations and are usually obligated to provide traceability to the private and public sector. Legacy GNSS installations are seldomly pre-calibrated upon installation and are most often not designed to be taken out of its context for calibration. In-situ GNSS time calibrations are either done by relative calibrations using a transported constant receiver system in common clock, or by transporting a clock of sufficient stability to provide the calibration of a local reference plane. However, such calibrations are often cumbersome if installations are not easily accessible. Relative calibrations require access to open sky conditions local to the site to be calibrated, which may be restricted or practically impossible to achieve even temporally. Clock transports are also posing risks and the achievable calibration uncertainty for long trips may not meet the user requirements. RISE has developed aiding techniques to allow calibrations in difficult environments, two of which are presented in this paper. At sites that have preinstalled dark fiber connectivity to a close to open sky situation, we deploy a White Rabbit link using BiDi optics to replicate the local reference plane at the traveling receiver setup. The used WR equipment introduces uncertainties in the order of a few hundred picoseconds in addition to the systematic uncertainties of the relative GNSS calibration, which is typically in the order of a few nanoseconds. The method helps to mitigate problems to access good receiving conditions of the outdoor calibration system and has a nominal range of up to 20 km. The outdoor part of the calibration system is self-contained regarding power for up to 15 hours of operation and can be freely placed. Differential real-time comparisons help to assess proper installation and to monitor its operation. In cases where a traveling receiver system cannot be connected, neither coaxially nor optically, the only viable option is a travelling clock. In order to minimize the requirements on that clock and to maximize calibration performance, we suggest a GNSS aided clock that is repeatably compared to the local reference plane using time interval measurements and is characterized in phase and frequency using a traveling GNSS receiver system placed ad hoc outdoors close to the site of calibration. Such scheme replaces a long trip with a running clock and makes a calibration more manageable and bound in uncertainty. The calibration uncertainty is partly dependent on the inherent stability of the transported clock in a changing environment and the time it takes to transport the clock to outdoor conditions for characterization. Typically, either a Rubidium or an OCXO are suitable and those must be chosen not to increase the overall calibration uncertainty. Additional uncertainty is of course caused by the GNSS observations and is dominated by the local environment around the temporarily placed antenna. A typical calibration would observe the transported clock for a period of about half an hour to one hour with the traveling GNSS system and then transport the clock to the local reference plane for time interval measurements for a few minutes of comparisons. The procedure is repeated several times in order to assure reproducibility and sound statistics of the calibration exercise, which can be achieved with a few nanoseconds uncertainty. In both cases the traveling equipment is as usual calibrated prior to the calibration trip and in closure after. Keywords: GNSS time calibration, White Rabbit, transportable clock, traceability