GPS Galileo Time Offset: How It Affects Positioning Accuracy and How to Cope with It

Kurzfassung

Looking forward for the advent of the future European navigation system Galileo, we anticipate a big part of its potential users to utilize combined GPS/Galileo receivers. It will allow such users to get benefits of observing two satellite constellations (= more satellites in view), and utilizing services of two independent navigation systems (= more reliability). The combined use of GPS and Galileo requires the systems to be interoperable. Already on the early stages of the Galileo program, interoperability with GPS was set as a goal for Galileo design and implementation. Leaving aside such issues like signal design, frequency allocation, signal processing etc, we concentrate here on the timing aspects of the interoperability. Like GPS today, Galileo will establish an independent timescale - Galileo System Time (GST) - which will serve as a basis for system operation and computation of satellite orbits and clock parameters. Both GPS Time and GST will be steered to the International Atomic Time (TAI). Their residual offsets to TAI will be available to users in the navigation messages. However, due to uncertainties of the TAI offset determination and of the steering, the accuracy with which the offset between GPS Time and GST will be known to users is expected to be in order of 10 ns. This uncertainty will cause a slowly changing bias between GPS and Galileo measurements in a combined navigation solution. The first part of the paper is dedicated to an estimation of the magnitude of the bias and its impact on the accuracy of user positioning which is taken as a function of the geometry of GPS and Galileo constellations (Dilution of Precision), equivalent user range error (UERE), and the uncertainty of the GPS Galileo time offset. With these three parameters we simulate the accuracy of the combined positioning solution for a world-wide grid, and empirically assess its dependence on the GPS Galileo time offset. The simulation allows us to obtain the upper limit for this offset at which its impact on the user positioning accuracy is below a selected significance level. The current Galileo baseline foresees the determination of the GPS Galileo time offset (GGTO) on system level and its dissemination to users in the Galileo navigation message. In the second section of the paper we discuss two options for the GGTO determination, namely: a) using a combined GPS/Galileo receiver, and b) using a GPS time receiver connected to the physical realisation of GST as generated at the Galileo Precise Time Facility (PTF). First, we assess the limits for accuracy of the GGTO determination set by the stability of GPS Time and GST, and propose a model which can be used to simulate the dynamic behaviour of GGTO. Then we study GGTO determination accuracy for both of the options listed above using simulated GPS and Galileo observations and the GGTO model we obtained. The simulations were made considering the accuracy of both today´s and modernized GPS. Finally, we check the simulation results by processing GPS measurements collected in the time laboratories of German Aerospace Center (DLR) and of the Royal Observatory of Belgium. In both processing tests, the timescale of the laboratory that hosts GPS receiver is assumed to represent GST. GPS data are processed with three types of satellite ephemeris - broadcast ephemeris, precise ephemeris computed by IGS, and broadcast ephemeris with correction from EGNOS (simulated) - to identify a candidate scenario for the operational implementation. Finally, we study the scenario where GGTO is determined in the user receiver as an additional unknown in the navigation solution. We simulate GPS and Galileo observations both for users with an unobstructed sky view and for users in a typical urban environment where the number of satellites is limited because of obstructions inherent to "cityscape". We simulate urban environments making use of known statistical models of cities implemented in DLR´s simulation software. These simulations allow to compare the two GGTO determination scenarios - on system level (broadcast of GGTO in the navigation message), and on user level (GGTO determination in user receiver) - in terms of user position accuracy and availability of the navigation service for the two types of users introduced above.