1,720,983 research outputs found
Time metrology in Global Navigation Satellite Systems
Get PDFPrecise timekeeping is at the basis of any Global Navigation Satellite System. In this thesis, after an extensive introduction on time and frequency metrology, some of the basic time-related aspects of navigation systems are discussed, and new ideas and solutions are presented. In the first part of the work, the most relevant innovative contributions are related to the mathematical clock model and to the stability analysis of atomic clocks affected by frequency jumps, as well as to the development of a new averaging algorithm for the generation of a robust time scale from an ensemble of atomic clocks. In the second part, devoted to the role of timekeeping in satellite navigation systems, the innovative contributions are mainly about: a revision of the relativistic corrections; the development and testing of a new composite clock, which could be used as a system time scale for the Galileo system; a study on the impact of the light-shift effect on the timing performance of GPS rubidium clocks; the development of a new recursive clock anomalies detector, as well as a discussion about the possible implementations of a clock anomalies detector and a compensation system for on-board applications
The Allan variance in the presence of a compound Poisson process modelling clock frequency jumps
No full textAtomic clocks can be affected by frequency jumps occurring at random times and with a random amplitude. The frequency jumps degrade the clock stability and this is captured by the Allan variance. In this work we assume that the random jumps can be modelled by a compound Poisson process, independent of the other stochastic and deterministic processes affecting the clock stability. Then, we derive the analytical expression of the Allan variance of a jumping clock. We find that the analytical Allan variance does not depend on the actual shape of the jumps amplitude distribution, but only on its first and second moments, and its final form is the same as for a clock with a random walk of frequency and a frequency drift. We conclude that the Allan variance cannot distinguish between a compound Poisson process and a Wiener process, hence it may not be sufficient to correctly identify the fundamental noise processes affecting a clock. The result is general and applicable to any oscillator, whose frequency is affected by a jump process with the described statistics
On-orbit GPS RAFS lamplight variations: statistics of lamplight jumps
No full textIn the rubidium atomic frequency standard (RAFS), an rf-discharge lamp produces the device’s atomic signal. As a consequence of the light-shift effect, variations in the lamplight’s intensity result in variations in the RAFS’ output frequency. While the basic physics of the light-shift is reasonably well understood, the temporal variations in the lamp’s light intensity are not. Here, we describe our analyses of lamplight intensity jumps occurring for the RAFS onboard GPS-IIR satellites. Briefly, we find that lamplight intensity jumps correspond to a compound Poisson process. The magnitude of the jumps appears to be a mean-zero random process with a family-wide standard deviation of 0.1%. The time between jumps corresponds to a Poisson process, and there appear to be two timescales associated with the jumps: a “fast” timescale, with a mean time between jumps of 7 days, and a “slow” timescale with a mean time between jumps of 180 days
Mitigation of lamplight-induced frequency jumps in space rubidium clocks
No full textRubidium clocks are currently the most common atomic clocks for space applications, playing a fundamental role in global navigation satellite systems. Their stability is affected by the light-shift effect, turning lamplight variations into frequency variations, e.g., lamplight intensity jumps into frequency jumps. In our previous work, analyzing data from GPS rubidium clocks, we uncovered the impact of the lamp on the in-orbit clock’s
performance. Specifically, the rubidium clock’s random walk of frequency seems to be driven by a compound Poisson process associated with lamplight intensity jumps. Most important, large lamplight-induced frequency jumps could affect the validity of the navigation message. Here, we propose and test on simulated data, a software compensation scheme for lamplight-induced rubidium clock frequency jumps. We show how this could be implemented as an automated onboard process, and the potential improvements this scheme might yield in timekeeping and navigation
performance. In particular, we demonstrate the possibility to correct large lamplight-induced frequency jumps in a time smaller than the interval between consecutive navigation message upgrades, thus improving the quality of the navigation message
Influence of the ac-Stark shift on GPS atomic clock timekeeping
No full textThe ac-Stark shift (or light shift) is a fundamental aspect of the field/atom interaction arising from virtual transitions between atomic states, and as Alfred Kastler noted, it is the real-photon counterpart of the Lamb shift. In the rubidium atomic frequency standards (RAFS) flying on Global Positioning System (GPS) satellites, it plays an important role as one of the major perturbations defining the RAFS' frequency: the rf-discharge lamp in the RAFS creates an atomic signal via optical pumping and simultaneously perturbs the atoms' ground-state hyperfine splitting via the light shift. Though the significance of the light shift has been known for decades, to date there has been no concrete evidence that it limits the performance of the high-quality RAFS flying on GPS satellites. Here, we show that the long-term frequency stability of GPS RAFS is primarily determined by the light shift as a consequence of stochastic jumps in lamplight intensity. Our results suggest three paths forward for improved GPS system timekeeping: (1) reduce the light-shift coefficient of the RAFS by careful control of the lamp's spectrum; (2) operate the lamp under conditions where lamplight jumps are not so pronounced; and (3) employ a light source for optical pumping that does not suffer pronounced light jumps (e.g., a diode laser)
Light-shift coefficient in GPS rubidium clocks: Estimation methods using lamplight/frequency correlations
No full textThe frequency of the Rubidium Atomic Frequency Standard (RAFS) used in Global Navigation Satellite Systems (GNSS) is affected by the light-shift effect, due to the opticalpumping light from the RAFS' rf-discharge lamp. As a consequence, lamplight intensity variations can induce RAFS output frequency variations, with lamplight stability setting a lower bound to RAFS frequency stability (i.e., to the eventual navigation performance of the GNSS). We study this effect by estimating the light-shift coefficient of an on-orbit RAFS and its possible variation in time using two different methods: the first makes use of large observed frequency jumps that are induced by lamplight jumps; the second uses the linear correlation between deterministic frequency and lamplight variations. We validate the methodology using GPS Block IIR RAFS data and present some preliminary results
- …
