Enhanced GNSS Satellite Clock Predictions - A Two-Stage Adaptive Approach

Ashkan Izadpanah, Mohamed Youssef

Abstract:	Uncertainty in the Global Navigation Satellites Systems (GNSS) satellite clock predictions is one of the key contributors to User Ranging Accuracy (URA), and consequently, user positioning accuracy. The accuracy of these predictions depends on the stability of the satellite clock itself and the prediction model used. Rx Networks Inc. performs GNSS satellite orbit and clock predictions as part of its Assisted-GNSS (A-GNSS) extended ephemeris products. The motivation for developing new satellite clock modeling techniques is to improve the overall accuracy performance of extended ephemeris solutions for GPS and GLONASS. Results show that this new Two-Stage Adaptive technique yields significant improvements (more than 2X in the clock predictions) into the clock predictions. In this paper, the technical approach taken to tackle this issue will be illustrated. The clock prediction enhancements are performed in two stages: 1) data preprocessing: detection and correction; and 2) modeling and prediction. The first data preprocessing stage is required to scan through the recorded historical real-time broadcast ephemeris database to check whether a clock anomaly has occurred and if so, then determine the type of the anomaly. Three major types of clock impairments were identified in the historical clock database: missing data, abrupt clock jumps, and clock trend changes. Missing data anomalies are corrected with a new interpolation-extrapolation [1] block designed to compensate for the data gaps in the database. Trend changes or clock jumps anomalies are corrected by removing the bad data from the historical database. After the clock anomalies are filtered in stage 1, the clean data is injected into a fitting tool in stage 2. The fitting tool uses a conventional Least-Squares (LS) technique to determine the optimal clock model and estimates its parameters based on the filtered historical data. Iterative techniques are also used since they prove to be more robust at handling clock outliers [2].
Published in:	Proceedings of the 24th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS 2011) September 20 - 23, 2011 Oregon Convention Center, Portland, Oregon Portland, OR
Pages:	2931 - 2936
Cite this article:	Izadpanah, Ashkan, Youssef, Mohamed, "Enhanced GNSS Satellite Clock Predictions - A Two-Stage Adaptive Approach," Proceedings of the 24th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS 2011), Portland, OR, September 2011, pp. 2931-2936.
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In