Observing the Behaviour of High Stable Galileo Satellite Clocks and Exploring Potential Associated Benefits

Abstract

It is quite common to consider GNSS satellites as highly accurate clocks orbiting the Earth and it is with the aid of these orbiting clocks that GNSS users are able to obtain precise positioning solutions. As for any other satellite navigation system, Galileo’s satellite clocks are one of the critical technologies in the system. In the particular case of Galileo, each satellite is equipped with four redundant clocks: two primary Passive Hydrogen Masers (PHM) measuring time to within 0.45 ns over 12 hours, which is four times better than the performance of the two secondary Rubidium clocks, accurate to within 1.8 ns over 12 hours, see [Ref. 1.]. Notwithstanding the reported and now identified Galileo satellite clock failures, see [Ref. 2.], the observed behaviour of the Galileo clocks, when operating nominally, seems to be very stable, especially in the short term. The clock stability feature is a key element: on the one hand for the accuracy of the navigation solutions and on the other hand for its non-negligible impact on the system complexity and associated costs. Clock stability, as well as the navigation solution accuracy, are directly related, together with other factors, to clock predictability. Higher stability implies more accurate predictability, and better predictions mean better ephemeris. Not only better standard ephemeris for navigation solutions, but also better long term ephemeris for assisted navigation. On top of that, better clock stability can also potentially contribute to reduce the complexity of the GNSS ground segment. Standard ODTS (Orbit Determination & Time Synchronisation) approaches implement snapshot strategies for the clock restitution, based on obtaining epoch by epoch estimates of all satellite’s and station receiver’s clock parameters. This approach implies the management of a large number of parameters in the estimation process, which could be drastically reduced if some information about the physical behaviour of the clocks in the system could be input to the processing filter, for example, if the snapshot estimation strategy could be replaced by a pure model or by a mixed model-snapshot strategy. This paper is aimed at observing the behaviour of the Galileo satellite apparent clocks, with the purpose of finding interesting features allowing potential improvements in at least the following aspects: • accuracy of the navigation solutions, • accuracy of long term ephemeris and • ground segment complexity. The results included in this paper are based on the analysis of apparent clocks obtained through accurate restitution by means of ODTS processes, implementing state-of-the-art algorithms and models. Typical trends and anomalous events of the Galileo satellite clocks are going to be described and analysed. Comparisons with other GNSS satellite clocks are going to be shown, and conclusions, driven out of the performed research, are going to be extracted.