Multi-Frequency Precise Point Positioning using GPS and Galileo Data with Smoothed Ionospheric Corrections
Francesco Basile, Terry Moore, Chris Hill, Nottingham Geospatial Institute, University of Nottingham, UK; Gary McGraw and Andrew Johnson, Rockwell Collins
Precise Point Positioning (PPP) is a carrier phase – based positioning technique that guarantees centimeter to decimeter level accuracy, with no need for reference stations. It can be considered as an improved version of the classic pseudorange – based positioning, in which broadcast navigation data are replaced with precise orbits and clocks estimates from a network solution. Models for environmental and site location effects are also required. The main corrections include tropospheric delay, antenna phase centre offset, ocean and solid Earth tides loadings, and antenna phase wind up effects. Using carrier phase measurements, in addition to code pseudoranges, means that the initial ambiguities have to be estimated, causing long convergence time. Typically, 20 to 40 minutes are required to have an accuracy of better than 10 centimetres. Its long convergence time represents the main drawback of PPP.
Traditionally in PPP, dual–frequency observations are combined into the ionosphere – free (IF) combination to eliminate the first order ionospheric delay or else ionospheric states are included in the PPP Kalman Filter with dual-frequency measurements provided to the filter. The latter approach has a significant computational cost compared to the iono-free approach, but has better solution re-convergence in transient signal environments since the PPP filter can retain information about the ionospheric delay. The computational penalty associated with the filter formulated with dual-frequency measurements and ionospheric states becomes more pronounced with increasing numbers of satellites associated with the use of multiple constellation, which is desirable for PPP in degraded signal environments.
The research reported in this paper concerns: 1) Assessment of the performance improvement offered dual-constellation (GPS/Galileo) PPP compared to single-constellation; 2) A federated iono-estimation and iono-free PPP technique that provides the rapid re-convergence provided by having ionospheric states in the PPP filter, with substantially reduced computational burden.
With the evolving GNSS landscape, the International GNSS Service (IGS), with its Multi – GNSS Experiment (MGEX), has made an effort to provide precise products for the new generation systems. Various Analysis Centres (ACs) are involved in the estimation of Final and Rapid orbits and clocks, as well as Differential Code Biases (DCBs), for GPS, GLONASS, Galileo, Beidou, and QZSS. However, these products can only be used in post – processing applications. At the moment, studying the PPP performance of the new systems for Real – Time applications is still limited by the lack of freely available Real – Time products. Indeed, the IGS Real – Time service only broadcast GPS and GLONASS products.
In this research, a simulator for Multi – Constellation Multi – Frequency (MCMF) GNSS observations and products was developed to study the performance of different signals combinations from GPS and Galileo, alone and together, in PPP in different environments. The model used to simulate the errors on the GPS and Galileo precise products is based on the comparison between the GPS Real – Time products provided by the IGS and the ESA Final products, which are considered as truth reference. This analysis showed that the error in the GPS Real – Time products have a periodic behaviour with two major component, one with period equal to the orbital period of the GPS constellation, and the other with a period that equals the Earth rotation period.
Availability, Dilution of Precision (DOP), accuracy, and convergence for GPS only, Galileo only and GPS plus Galileo are compared in both open sky and constrained and transient signal environments associated with urban environments. The latter is particularly interesting to test the re-convergence time of PPP. It is well known that, in case the GNSS receiver loses track of the carrier phases, the positioning filter needs to be reinitialized, meaning that further tens of minutes are required before re-convergence.
Results show better positioning performance of Galileo signals as compared to GPS signals. While, using the two systems together provides improvements particularly in the convergence time. This paper will also consider the specific impact of the Galileo Alternative BOC (AltBOC) modulation. Even though Galileo E5 is less noisy and provides better multipath rejection than E5a, this research demonstrates that they have the same positioning performance due to the influence of E1 signal errors in the IF combination.
Finally, a new method to mitigate the ionosphere delay was proposed in order to ensure the best positioning performance from multi – frequency PPP. Instead of using the IF combination, here the uncombined code pseudoranges are corrected with ionosphere delay information coming from federated carrier smoothing (Hatch) filters for each satellite. This new method provides faster re-convergence time and improved positioning accuracy.