Combination of multi-GNSS products

The International GNSS Service (IGS) is well established as a scientific reference for GPS data processing since more than 20 years. The observations from a global tracking network are processed by a number of Analysis Centers (ACs, typically around ten nowadays) and individual ACs’ results are combined to the official IGS products by the Analysis Center Coordinator (ACC) using an algorithm described by Beutler et al. (1995).

Some of the ACs have extended their activities to GLONASS by a rigorously combined GPS and GLONASS solution. At the same time, the combination within the IGS takes place in two fully independent chains for GPS and GLONASS.

Extrapolating the current situation by the current development undertaken in the frame of IGS-MGEX (Multi-GNSS Experiment) we will be faced in the future with inhomogeneous contributions from the different IGS ACs:

GPS
us-icon
GLONASS
ru-icon
Galileo
eu-icon
BeiDou
cn-icon
QZSS
jp-icon
IRNSS
in-icon
AC1
AC2
AC3
AC4
AC5
(Hypothetical scenario on the selection of constellations by IGS ACs)

On top of this, with the modernization programs of the established Global Navigation Satellite Systems (GNSS) and the new systems under construction, the manifold of frequencies and signals will significantly increase. This development is accomplished by an even more dramatic increase of signal tracking techniques indicated, which likely will increase the variability between the AC’s contribution.

Emerging questions:

  • What is the optimal combination strategy in order to:
    • keep the internal consistency between the systems from the contributing solutions into the combined products?
    • obtain the optimal quality for the combined orbits of all satellites considering that the quality of the orbits from the different constellations may vary?
  • In particular for the clock combination:
    • How to consider the potential biases if each AC shall be free to choose any of the frequencies, signals and available observation types?

In order to answer these questions the Astronomical Institute of the University of Bern (AIUB) and the European Space Operations Centre (ESOC) have established a PhD work in the frame of the Networking Partnership Initiative (NPI) program.

Results have for instance been presented in: