Estimating the Allan variance from frequency measurements with missing data

The Allan variance is the key measure for stability analysis, a fundamental tool to establish the performances of atomic clocks, which are often critical elements in many applications, such as global navigation satellite systems (GNSS). In some applications, only frequency measurements with numerous missing data are available. Because of missing data, the resulting Allan variance can be dramatically different from the expected stability behavior. We have developed an Allan variance estimator which corrects this divergence. We show the effectiveness of our estimator by applying it to frequency measurements with up to 94.4% of missing data.

[1]  Lorenzo Galleani,et al.  Application of the Dynamic Allan Variance for the Characterization of Space Clock Behavior , 2011, IEEE Transactions on Aerospace and Electronic Systems.

[2]  D. W. Allan,et al.  Statistics of atomic frequency standards , 1966 .

[3]  Lorenzo Galleani,et al.  The dynamic Allan Variance III: Confidence and detection surfaces , 2011, IEEE Transactions on Ultrasonics, Ferroelectrics and Frequency Control.

[4]  Measures for random instabilities in frequency and time (phase) , 2017 .

[5]  Patrizia Tavella,et al.  Statistical and mathematical tools for atomic clocks , 2008 .

[6]  Lorenzo Galleani,et al.  Inter-satellite links for satellite autonomous integrity monitoring , 2011 .

[7]  D. W. Allan,et al.  Time and Frequency (Time-Domain) Characterization, Estimation, and Prediction of Precision Clocks and Oscillators , 1987, IEEE Transactions on Ultrasonics, Ferroelectrics and Frequency Control.

[8]  I. Sesia,et al.  Estimating the Allan variance in the presence of long periods of missing data and outliers , 2008 .

[9]  P. Tavella,et al.  Time and the Kalman Filter , 2010, IEEE Control Systems.

[10]  C. Salomon,et al.  Atomic clock ensemble in space , 2011 .

[11]  P. Tavella,et al.  The dynamic Allan variance , 2009, IEEE Transactions on Ultrasonics, Ferroelectrics and Frequency Control.

[12]  G. Mana,et al.  Use of Bayesian statistics to reduce the density shift uncertainty in Cesium fountain , 2010, 2010 IEEE International Frequency Control Symposium.

[13]  Lorenzo Galleani,et al.  The dynamic Allan Variance IV: characterization of atomic clock anomalies , 2015, IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control.

[14]  Cristina Zucca,et al.  A mathematical model for the atomic clock error in case of jumps , 2015 .

[15]  L. Galleani,et al.  The dynamic Allan variance II: a fast computational algorithm , 2010, IEEE Transactions on Ultrasonics, Ferroelectrics and Frequency Control.