Multi-GNSS Precise Point Positioning

January 8th, 2015 by

We have just released magicGNSS version 5.5 which incorporates a new version of the Precise Point Positioning algorithm (PPP). The new PPP is able to process GPS, GLONASS and GALILEO.

Multi-GNSS precise satellite orbits and clocks needed as input by PPP are computed beforehand using magicGNSS‘ ODTS module. Multi-GNSS orbits and clocks are available starting January 1, 2014, with a latency of around 1 day. The new PPP has the following processing modes: GPS-only, GLONASS-only, GPS+GLONASS, GALILEO-only, GPS+GALILEO, and GPS+GALILEO+GLONASS. It is also possible to use the PPP service by email.

magicGNSS/PPP with MULTI-GNSS support  was presented and demonstrated at the ION GNSS 2014 conference in Tampa, Florida, USA. The title of our paper was “magicGNSS‘ Real-Time POD and PPP MULTI-GNSS Service”.

About station icons

January 8th, 2015 by

Just a quick explanation about the meaning of station icons in magicGNSS:

GPS+GLONASS core station

GPS core station

deselectedDeselected station

reference_clockReference clock (in ODTS)

GPS+GLONASS user station with precise coordinates

GPS+GLONASS user station (or rover) with approximate coordinates

GPS user station with precise coordinates

GPS user station (or rover) with approximate coordinates

GPS+GLONASS public station with precise coordinates

GPS+GLONASS public station with approximate coordinates

GPS public station with precise coordinates

GPS public station with approximate coordinates

Multi-GNSS core station

Multi-GNSS user station with precise coordinates

Multi-GNSS user station with approximate coordinates

First Galileo-only PPP with IOV + FOC-1 satellite

December 11th, 2014 by Guillermo Tobías

It has been over one week now since Galileo FM5 started transmitting. The early results obtained with magicGNSS when processing the data collected by IGS’ MGEX network ( during November 29th and 30th were quite preliminar, as they showed what seemed to be corrections being applied to the on-board atomic clock.

The results obtained when processing more recent data (from December 5th to December 9th, as it can be seen in the figure below), show a much more stable clock behaviour for FM5 satellite, which may lead to conclude that the apparent clock corrections have been successfully performed.

Based on this assumption, 2 consecutive 4-day long ODTS arcs have been executed with a 24 hour delay between them, and the overlap for the Galileo satellites, both for the orbit and the clock estimates, has been computed. The results presented below show that the orbit and clock consistency for all the Galileo satellites (both the IOV and the FM5), are in the same order of magnitude.

Galileo clocks consistency:

Galileo orbit consistency:

Despite the fact that the orbit and clock determination accuracy for FM5 can still be improved given the comparison with respect to the results obtained for the IOV satellites, the estimates can be considered accurate enough for thinking about using them for High Accuracy Positioning purposes.

In this regard, a Galileo-only PPP (3 IOV satellites transmitting in more that one frequency + FM5) has been run for LLAG station using as reference products the aforementioned estimates.

This GPS+Galileo MGEX station is located in the Canary Islands, and during December 6th the 3 usable IOV satellites ( PRNs E11, E12 and E19) and FM5 (PRN E18) satellite were visible over that station for around 2 hours, which allowed the computation of a Galileo-only PPP solution for LLAG station.

The estimated station coordinates were compared with the reference ones, which had been previously computed by means of a 24-hour long batch PPP using IGS reference products. The comparison shows that the Galileo-only PPP solution error is at the centimetric level. However, these results are quite optimistic due to the fact that the PPP estimated a single position based on a 2-hour data batch. In this regard we intend to perform a sequential Galileo-only PPP in order to obtain more realistic results about the positioning performances that could be achieved in a Real-Time environment.

Although this analysis can still be understood as a proof-of-concept, the presented results show the feasibility for using E18 satellite for High Accuracy, which can be considered as great news for the GNSS community!

Let’s hope that FM6 satellite follows the same path as FM5 so that it brings benefit for the GNSS community and in particular for High Accuracy purposes.

Early processing of first Galileo FOC satellite by magicGNSS

December 1st, 2014 by Guillermo Tobías

As published by ESA back on November 10th (, Galileo satellite FM5 has been repositioned during the month of November into a more circular orbit, raising its perigee and reducing its eccentricity from an initial value of 0.23 to a target one of around 0.156 (as stated during last ION GNSS 2014 meeting by ESA).

Last Saturday, E18 satellite started transmitting (as advanced by @GNSS_news) and IGS’ MGEX network ( is already tracking it. Based on the collected measurements during November 29th and 30th, and initial orbit and clock estimation for E18 has been performed by means of magicGNSS.

The 4 different clock intervals are shown below:

November 29th 06:15:00 to  November 29th 10:10:00:

After removing quadratic model adjustment:

November 30th 07:45:00 to  November 30th 12:25:00

After removing quadratic model adjustment:

November 30th 13:15:00 to  November 30th 19:25:00

After removing quadratic model adjustment:

November 30th 20:40:00 to  November 30th 23:55:00

After removing quadratic model adjustment:

We hope to better characterise E18 orbit over the next days as more data is available.

Broadcast of magicGNSS’ reference products in the Galileo CS signal

September 29th, 2014 by magicGNSS Team

During the test campaign of the EPOC (Early Proof-Of-Concept) in the frame of the AALECS (Authentic and Accurate Location Experimentation with the Commercial Service) project, magicGNSS reference products have been broadcast in the Galileo E6 signals.

Orbit and clock predictions have been generated 2-3 days in advance of each test by magicODTS using data gathered by the IGS MGEX network ( This high-accuracy (HA) data is then sent to GMS and uploaded to the IOV satellites for broadcast during test slots (see figure below).

The E6-B signal has been tracked both with SCE (Spreading Code Encryption) enabled and disabled. During tests broadcast products are gathered and authenticated via TESLA-based NMA and/or SCE. Then, a high-precision data-authenticated PVT solution is computed by means of magicPPP.

The aim of the test campaign has been to evaluate the E6 data transmission performance, and prove that the system is capable of providing authentication and HA positioning/timing services. Several static and kinematic tests have been conducted from mid-July to late-September 2014. The figure below shows performance of a static PVT with SCE enabled: an RMS below 40 cm in the horizontal plane is achieved, including the convergence period.