|Abstract:||1. Introduction The DLR and GFZ joint project ADVANTAGE (Advanced Technology for Navigation and Geodesy 1,2 ) aims at studying the potential of optical ranging and time transfer technology for future satellite navigation systems. The proposed constellation consists of 24 Medium Earth Orbit (MEO) navigation satellites and 4 Low Earth Orbit (LEO) satellites devoted to generate the time scale and synchronize the clocks of the system using two-way optical links. The MEO constellation is essentially similar to the current GALILEO system which assures long term orbit stability and limits the number of constellation keeping maneuvers. The LEO part consists of satellites on circular polar orbits on two perpendicular orbital planes (two LEOs per plane) with a mutual phasing that assures the LEO-MEO links to be optimal for the synchronization of the system time. Using GFZ’s Earth Parameters and Orbit System – Orbit Computation (EPOS-OC) software package we simulate two data types which will be used in precise orbit determination (POD): (1) GNSS type L-band ranging data (pseudo-range and carrier phase) between both, MEOs and LEOs (satellite-to-satellite tracking data) and MEOs and a network of 124 ground stations, (2) highly precise ranges between neighboring MEOs as well as between MEO and LEO satellites. 2. Objectives The main goal of this research is to find the impact of inter-satellite links and ultra-stable time on POD of the proposed constellation. In particular we want to asses the benefit of the inclusion of the LEOs for POD of the navigation satellites and to investigate possibilities of reducing the number of ground stations required to achieve satisfactory orbit accuracies. A realistic simulation must also include systematic modeling errors like solar radiation pressure, antenna thrust and phase center offset etc. We simulate them to find their influence on the performance of the recovered orbits and probable impacts on their use in geodetic applications like gravity field determination, remote sensing applications, Earth observation (water vapor, reflectometry) and precise point positioning. 3. Simulation In the data simulation step we use standards and models from daily processing of real GNSS data. That includes GPS-like attitude, ROCK4 solar radiation pressure model, EIGEN-6C gravity field model, Earth and ocean tide models, ocean loading etc. The antenna thrust is simulated as constant radial acceleration with small periodic variations. The GNSS range data are simulated on 2 GALILEO frequencies E1 and E5. For the purpose of orbit determination ionosphere-free combinations are established thereof and endowed with Gaussian noise of 50 cm for pseudo-ranges as well as 3 and 5 mm for carrier phases of LEOs and MEOs, respectively. To obtain a realistic number of carrier phase ambiguities, cycle slips are simulated as well. The system time is assumed to be ultra-stable for all the MEO and LEO satellites. Therefore the clock offsets are simulated as constant values with zero mean and small Gaussian noise of 0.3 mm. 4. Orbit recovery In the POD step we use the integrated approach. Here the simulated data from all satellites, MEOs and LEOs, and from all ground stations are processed simultaneously. The parameters of interest, like the orbits, are estimated in a differential orbit improvement process following the least-squares (LS) principle. To achieve the goals of this research, we define various orbit recovery scenarios involving different combinations of MEOs, LEOs and the ground network with different data types and a selection of modeling errors. In particular, in view of the proposed constellation we consider POD using: a) MEOs + LEOs only, b) MEOs + LEOS + a limited number of ground stations. For each constellation set-up we investigate the impact of modeling errors, ambiguity fixing and inclusion of precise inter-satellite range links. One of the most important properties of the proposed navigation system is the very stable timekeeping, which results in predictable clock offsets of the space segment. To asses the benefit of stable clocks, we use three methods to obtain the clocks. In the first one all clock offsets are estimated epoch by epoch, which is the standard case in current day GNSS data processing. A known drawback is the necessity to estimate a large number of clock parameters. In the second approach we make use of ultra-stable clocks and model them as being piece-wise constant. The ensuring significant reduction of the number of adjustable parameters makes the LS solution more accurate and more reliable on the one hand. On the other hand the modeling errors become more visible in observation residuals, orbits and other parameters, as they are not absorbed by the large number of clock parameters as in the first method. In the third approach we consider a theoretical case when all space clocks are known and can be fixed. Finally we summarize the results with focus on the orbit accuracy. Footnotes: 1. This project is supported by the Helmholtz-Gemeinschaft Deutscher Forschungszentren e.V. under grant numbers ZT-0007 (ADVANTAGE, Advanced Technologies for Navigation and Geodesy). 2. The project ADVANTAGE is a joint project of DLR and the GFZ German Research Centre for Geosciences. It aims at defining a future system for navigation, geodesy and metrology.|
Proceedings of the 31st International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2018)
September 24 - 28, 2018
Hyatt Regency Miami
|Pages:||968 - 1001|
|Cite this article:||
Michalak, Grzegorz, Neumayer, Karl Hans, Koenig, Rolf, "Precise Orbit Determination with Inter-satellite Links and Ultra-stable Time for a Future Satellite Navigation System," Proceedings of the 31st International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2018), Miami, Florida, September 2018, pp. 968-1001.
ION Members/Non-Members: 1 Download Credit