Testing the robustness of a time scale algorithm by using simulated optical clock data

Optical clocks have reached an impressive accuracy, surpassing by orders of magnitude that of microwave clocks, so that the optical transitions on which they are based can be considered as ideal candidates for a possible future redefinition of the second. For these reasons, it is fundamental to develop an effective way to generate time scales based on optical clocks. However, optical clocks are still far from matching the robustness, reliability and long times of autonomous operation typical of microwave clocks. Therefore, the usual way to generate a time scale based on an optical clock, is to use it as a frequency steering reference for a master clock [1], typically an Active Hydrogen Maser (AHM). The master clock is usually referred to as a flywheel oscillator, since it allows the generation of the time scale also when the optical clock is unavailable. It is hence important to understand the impact of the unavailability of the optical clock on the performances of the steering algorithm and of the generated time scale. For example, it is fundamental to understand the minimum availability of the optical clock data needed to guarantee a given level of performances. To this aim, we simulated a time scale steered with an optical clock by considering several possible scenarios for the availability of the latter. By modifying the steering algorithm described in [2], we simulated the measurement data of the frequency offset of an AHM with respect to an optical clock, and we used them to steer the AHM for the time scale generation. Then, we tested the algorithm by considering six different scenarios for the availability of the optical clock, spanning from the ideal one (continuous operation of the optical clock), to the worst one (non-uniformly distributed frequency measurements with long unavailability periods). We also considered a realistic scenario where the optical clock is operated for a few hours once a week, with the possibility of a jitter on the day of the week. Finally, we evaluated the performances by considering the phase offset of the steered time scales with respect to UTC, and we compared and discussed the results obtained within the different scenarios. The results prove that the steering algorithm is robust and effective despite its very simple implementation. As expected, the scenarios giving the best and worst performances are the ideal one and those with long unavailability periods, respectively. More interestingly, the realistic scenario, with one measurement per week only, gives results similar to the ideal scenario ones. This is remarkable, as it means that, even with a much smaller effort dedicated to the optical clock, the final performances of the time scale are still close to the optimal case. This project 18SIB05 ROCIT has received funding from the EMPIR programme co-financed by the Participating States and from the European Union’s Horizon 2020 research and innovation programme. References [1] H. Hachisu, F. Nakagawa, Y. Hanado and T. Ido, “Months-long real-time generation of a time scale based on an optical clock,” Scientific Reports, (2018) 8:4243. [2] L. Galleani, G. Signorile, V. Formichella and I. Sesia, “Generating a real-time time scale making full use of the available frequency standards,” Metrologia 2020, in press.