magicSBAS is integrated into the IGS Real Time Service
The GMV proprietary SBAS testbed, magicSBAS, has been successfully upgraded to connect to RTCM real time service (RTS) providers broadcasting precise satellite orbit and clock corrections. The integration and performance tests have been conducted with both a server of the magicGNSS suite and with the International GNSS Service (IGS) RTS IGS01/IGC01 server, which was launched recently on April 1st.
magicSBAS is a state-of-the-art, multi-constellation, operational Satellite Based Augmentation System (SBAS) testbed developed by GMV to offer SBAS regional differential corrections and non-safety critical integrity augmentation to any interested region, and it is being extensively used in different SBAS programs worldwide for engineering and demonstration purposes.
When connected to RTS, magicSBAS uses the external orbit and clock corrections instead of its internal algorithms to generate the SBAS satellite long term and fast messages. The correction bounds, the UDREs, are still computed internally with real time measurements from a regional network of reference stations to ensure integrity within the demanding SBAS time to alert. Special care is paid to the estimation of the offsets between the different time scales used by the external provider and the SBAS solution and to refer each satellite clock to C/A code, as requested in SBAS L1. This new execution mode of magicSBAS has been shown to double the accuracy of the satellite ranging error at worst user locations (SREW) within its service area.
Integration with RTS is not the only improvement in the new version of magicSBAS. In addition to enhanced robustness and better connectivity to combined sources of real time raw measurements data, such as NTRIP and EGNOS EDAS raw data, the tool upgrade can be configured to use dynamic PRN masks. The SBAS L1 standard limits the number of satellites to be augmented simultaneously to 51. Where two or more GNSS core constellations are to be covered, it would be necessary either to preselect a subset of satellites or to change the PRN mask at regular intervals to augment a different set of satellites, a technique known as “dynamic PRN mask”. For the first time ever, a real time SBAS testbed implements such a functionality. The mask is modified at rates in the range of one hour, optimizing the performance in the worst user location of the service area. The initial tests run with GPS and GLONASS constellations show outstanding performance in the number of monitored satellites and vertical dilution of precision at the worst user location. For instance, the minimum number of monitored satellites at the worst user of a list of representative locations throughout the European continent rises from 8 to 12.
This recent evolution of magicSBAS is the result of GMV’s determined commitment to innovation in GNSS, and more particularly in SBAS data processing.
Figures:
Fig. 1: Results of integration of magicSBAS with IGS Real Time Service
Nominal SREW histogram (meters). Real time data from EGNOS EDAS reference station network (17/04/2013)
SREW histogram (meters) with external orbits and clocks. Real time data from IGS01 RTS stream and EGNOS EDAS reference station network (17/04/2013).
Fig. 2: Static Vs. Dynamic PRN masks. Histograms comparing the number of monitored satellites (left) and VDOP (right) for static PRN mask (top) and dynamic PRN mask (bottom). At each one second epoch, the number of satellites and the VDOP were selected as the worst cases of 20 user locations. In the dynamic mask case, 30 GPS and 24 GLONASS were configured. In the static mask case, three GPS satellites (PRNs 22, 23 and 24) and three GLONASS (slot numbers 22, 23 and 24) were excluded with respect to the dynamic configuration. The experiment was carried out with one-and-a-half day’s real data from April 2012.
# of monitored satellites VDOP