Towards an Optimal Strategy for GPS Wet Delay Kalman Filtering

Torben Schueler, Guenter W. Hein and Bernd Eissfeller

Abstract:	Troposphere parameters like zenith wet delays which are of high interest for climate research and meteorology can be estimated using GPS phase measurements. The technique of Kalman filtering has proven to be a suitable and efficient means for this purpose. However, several problems have to be addressed. One major problem is the proper definition of the underlying stochastic behavior of the wet delays. Normally, zenith wet delays are treated as random walk stochastic processes. Process noise parameters from the analysis of GPS-derived troposphere delay time series for a number of globally distributed IGS tracking stations are presented in this paper. The results suggest values of about 2 to 9 mm/Öh depending on the location of the station. Time series analysis provides average values for the process noise. However, wet delay variations occurring within a particular time window can be rather different and to account for such deviations from the mean behavior the concept of dynamic process noise has been implemented in the institute’s tropospheric analysis software TropAC. Instead of defining an average process noise value for the whole diurnal data batch, hourly process noise values are derived from the wet delays either by applying models or by a priori knowledge of the delays. The latter gives best results but requires an extra iteration as the variability analysis has to be carried out by using results from the initial filter run.
Published in:	Proceedings of the IAIN World Congress and the 56th Annual Meeting of The Institute of Navigation (2000) June 26 - 28, 2000 The Catamaran Resort Hotel San Diego, CA
Pages:	762 - 771
Cite this article:	Schueler, Torben, Hein, Guenter W., Eissfeller, Bernd, "Towards an Optimal Strategy for GPS Wet Delay Kalman Filtering," Proceedings of the IAIN World Congress and the 56th Annual Meeting of The Institute of Navigation (2000), San Diego, CA, June 2000, pp. 762-771.
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In