Improving the Performance of Multi-GNSS Time and Frequency Transfer Using Robust Helmert Variance Component Estimation

Abstract

The combination of multiple Global Navigation Satellite Systems (GNSSs) may improve the performance of time and frequency transfers by increasing the number of available satellites and improving the time dilution of precision. However, the receiver clock estimation is easily affected by the inappropriate weight of multi-GNSSs due to the different characteristics of individual GNSS signals as well as the outliers from observations. Thus, we utilised a robust Helmert variance component estimation (RVCE) approach to determine the appropriate weights of different GNSS observations, and to control for the influence of outliers in these observation in multi-GNSS time and frequency transfer. In order to validate the effectiveness of this approach, four time links were employed. Compared to traditional solutions, the mean improvement of smoothed residuals is 3.43% using the RVCE approach. With respect to the frequency stability of the time links, the RVCE solution outperforms the traditional solution, particularly in the short-term, and the mean improvement is markedly high at 14.89%.