Implementation of UERE Module into NavSim and Study of Errors for Single and Dual Frequency GPS-Receivers

Kurzfassung

In car, train or air navigation, in precision farming, in geodesy or in survey, the level of performance required differs. Under performance, it is often meant accuracy. One of the most important performance parameters that characterize the accuracy of a GNSS (Global _Navigation Satellite Systems) is the UERE (User Equivalent Range Error). This parameter coupled with the geometry of visible satellites gives the information of the mean accuracy a user can reach. The aim of this work is to study two types of receivers: The single frequency receiver and the dual frequency receiver. A former work [2] performed the study of a single frequency receiver by using MATLAB. The specificity of this work was to implement the error model of a single frequency receiver into NavSim (the end-to-end navigation simulator developed by DLR) [6]. An extension to a dual frequency receiver has been performed directly in NavSim and results of this study have been obtained and discussed. The general concept of GNSS errors is derived from the observation equation expended to a first order Taylor series [1] [2]. From a single frequency observation equation (L1) and the so-called “ionosphere free” observation equation on L1 and L2, it is possible to derive the fundamental error equation using the concept above. This master’s thesis report describes how the UERE measurement algorithms have been implemented into NavSim by the development of one new optional module “UERE”. This module uses post processed files of the International GNSS Service (IGS) (SP3, RINEX, IONEX, navigation file etc.) as reference files to calculate the deviation from an estimated quantity as a receiver would process it. At a particular time epoch, the difference between reference and estimation gives the Instantaneous Pseudo Range Error (IPRE) at both single frequency (L1) and double frequency (L1 and L2). The module stores all the errors and IPRE sequentially in the corresponding output text files. A statistical analysis of the output files is documented in the thesis report. As expected, one observes the dominant error of IPRE is the ionospheric error at single frequency. It is shown that this error can vary significantly from one year to another, and one location to another. Clock and ephemeris errors also show some regular behaviour or correlations that are specific for each satellite over time. In the case of dual frequency, ionospheric free combination of multipath and noise together gives reduced error in the IPRE level (because of exclusion of ionospheric error from IPRE), but the multipath and noise increases compare to the single frequency. For some locations we noticed correlation between ionospheric and tropospheric errors. Correlation studies between multipath and noise observables at L1 and L2 frequency shows that they are not correlated. Finally, some future work for further analysis of the results is proposed.