Skip to main content Accessibility help

High Spatial Variation Tropospheric Model for GPS-Data Simulation

  • Ashraf Farah (a1), Terry Moore (a1) and Chris J Hill (a1)


Precise GPS simulated data requires accurate simulation of the two major sources of error in GPS measurements, namely the ionospheric and tropospheric delays. The ionospheric delay modelling has been handled in a previous work (Farah, 2002). In this paper the simulation of the tropospheric delay is discussed. The suggested model should be accurate in estimating the tropospheric delay as well as capable of simulating high spatial variations of the troposphere resulting in more realistic simulated GPS data. In this paper, the EGNOS tropospheric correction model is considered as a possible tool for simulating the tropospheric delay in order to obtain more realistic simulated GPS data. Comparing the total tropospheric zenith delays from the EGNOS model with the CODE-tropospheric product has allowed the quality of the EGNOS model to be assessed. Four IGS-tracking stations have been selected for this study. Data from four non-consecutive weeks in different seasons over a period of one year were tested to assess the seasonal variation of the weather conditions. It is shown that the EGNOS model agrees well with the CODE-estimations with a mean zenith delay difference of approximately 2 cm. The maximum zenith delay difference between the EGNOS model and the CODE-estimations was in the range of 5 cm to 16 cm, which agrees well with previous studies. A second study has investigated the behaviour of the EGNOS model with other established tropospheric models such as the Saastamoinen, the Hopfield, the Marini and the Magnet model for three IGS-stations. It can be concluded from this study that the EGNOS model shows better agreement with the IGS estimations than the Magnet model and compares well with other models. The major shortcoming in the EGNOS model is its inability to simulate the variations in the troposphere over small regions. This shortcoming could be overcome by using the theory of Gaussian Random Fields, which has been previously used to model real life phenomena such as surface roughness (Chan, 1999). This paper was first presented at ION GPS 2003, the 16th Technical Meeting of the Satellite Division of the Institute of Navigation held at Portland Oregon, USA.




Full text views

Total number of HTML views: 0
Total number of PDF views: 0 *
Loading metrics...

Abstract views

Total abstract views: 0 *
Loading metrics...

* Views captured on Cambridge Core between <date>. This data will be updated every 24 hours.

Usage data cannot currently be displayed