Long-term performance analysis of GIOVE clocks

Passive Hydrogen Maser (PHM) and Rubidium Atomic Frequency Standard (RAFS) are the two baseline on-board clock echnologies for Galileo, the European Global Navigation Satellite System (GNSS), which are currently being validated on-board two experimental spacecrafts, GIOVE-A and GIOVE-B. These two spacecraft are orbiting in a Galileo-representative orbit and transmit Galileo-representative signals that are tracked and collected by a network of evenly distributed sensor stations (the Galileo Experimental Sensor Stations). The collected observables (pseudo-range and carrier phase) are being processed by the GIOVE Processing Centre, providing, as one of its core products, the estimation of the phase offset between the on-board clock and a reference clock, chosen to be an Active Hydrogen Maser connected to one of the sensor stations. RAFS technology has been operated in-orbit for almost 5 years on-board GIOVE-A and PHM technology for more than 2½ years on-board GIOVE-B. The data accumulated over this period have been analyzed together with the long-term performances of these on-board clocks. On GIOVE-B, it will be shown that the GIOVE estimation noise limits the actual characterization of the PHM stability. Over the medium term (12~24 hours), the performance of the on-board clock is affected by oscillations at the orbital period whose possible causes will be discussed. Over the long term (> 1 day), the PHM exhibits excellent frequency drift performances (below 1×10-15/day). On GIOVE-A, it will be shown that the short-term stability of RAFS is not limited by the estimation noise and is below the specified limits (5×10-12/SQRT(tau)). Over the medium term, the stability is also affected by periodic oscillations at the orbital period that are mostly due to on-board thermal variations. Finally, over the long term, even if not always monotonic, the RAFS frequency drift is below the 1×10-13/day level.