Relations Between Generalized JST Algorithm and Kalman Filtering Algorithm for Time Scale Generation

TL;DR

This paper explores the mathematical relationship between a generalized JST algorithm and Kalman filtering in atomic time scale generation, revealing conditions for their equivalence and differences in residual variances.

Contribution

It provides a rigorous mathematical analysis linking the generalized JST algorithm with Kalman filtering, including conditions for their equivalence and variance comparisons.

Findings

Averaged atomic time in generalized JST does not depend on observation noise.

Equivalence between JST and Kalman filter occurs under specific covariance and weighting conditions.

Numerical examples demonstrate the theoretical results' practical implications.

Abstract

In this paper, we present a generalized Japan Standard Time algorithm (JST-algo) for higher-order atomic clock ensembles and mathematically clarify the relations of the (generalized) JST-algo and the conventional Kalman filtering algorithm (CKF-algo) in the averaged atomic time and the clock residuals for time scale generation. In particular, we reveal the fact that the averaged atomic time of the generalized JST-algo does not depend on the observation noise even though the measurement signal is not filtered in the algorithm. Furthermore, the prediction error of CKF- algo is rigorously shown by using the prediction error regarding an observable state space. It is mathematically shown that when the covariance matrices of system noises are identical for all atomic clocks, considering equal averaging weights for the clocks is a necessary and sufficient condition to ensure equivalence between the generalized JST-algo and CKF-algo in averaged atomic time. In such homogeneous systems, a necessary and sufficient condition for observation noises is presented to determine which algorithm can generate the clock residuals with smaller variances. A couple of numerical examples comparing the generalized JST-algo and CKF-algo are provided to illustrate the efficacy of the results.