GNSS Bias Calibration Process and Results

P. D’Angelo, J. Pulido, D. Rincón, P. Silva, P. Vieira, F. Amarillo

Abstract:	BACKGROUND The signal in space bias among two navigation signals (both transmitted by the same satellite) can be defined as the difference in the apparent clock between the two signals induced by the transmitter. It is caused by the electronic equipment generating the signals, different for each channel, inducing a different hardware delay in each frequency. The use of GNSS receivers onboard Low Earth Orbit (LEO) satellites is nowadays very common, given the excellent orbit determination performances that can be achieved. In case of relatively high LEO orbits, the measurements gathered by the receivers are mostly unaffected by the ionosphere, therefore representing a very especial source of information about SISBAF (Signal In Space Biases Amongst different Frequencies). If multi-frequency measurements from these missions are available, the estimation of SISBAF and TEC (Total Electron Content) can be de-coupled. On a first step, SISBAF can be estimated from these LEO measurements, achieving an accuracy of a few centimetres. On a second step, TEC values are estimated from the measurements gathered by the GNSS ground stations network, taking advantage of the SISBAF parameters estimated on the first step. INTRODUCTION The GNSS-BICS (GNSS Bias Calibration System) is designed to provide, under demand, the GNSS Medium Earth Orbit (MEO) S/C SISBAF, from GNSS observations gathered by GNSS multi-frequency receivers on-board LEO satellites (from different missions). It establishes a passive, transparent and indirect collaboration between two or more independent space systems; namely a GNSS system, based on MEO satellites, and other or others system(s) based on LEO satellites (from different missions). This collaboration does not require the modification of any of the individual space systems, but the development of a complementary, independent and limited ground infrastructure, estimating in pseudo-real-time the GNSS MEO S/C SISBAF. The concept has been developed during the GNSS-BICS project, being carried out by DEIMOS Space (Spain) and DEIMOS Engenharia (Portugal) in the frame of the ESA’s General Studies Programme (GSP). Preliminary system definition and assessment of the system concept have been described in a previous ION publication [1]. Objective of this paper is now to present the details of the function for SISBAF calibration and the results of the system validation campaign. Both real and synthetic input data are used to validate the GNSS-BICS processing schemes. The former obtained from existing LEO satellites carrying GNSS dual-frequency receivers, the latter generated by simulations. CALIBRATION PROCESS Real data are analysed with the objectives of identifying periods where the observables are free of any detectable ionosphere effect and the existence of harmonic behaviour. Multipath linear combination is used to evaluate the antenna effect variation with nadir angle as seen from each GNSS satellite to generate a table associating GNSS antenna correction per frequency for each Line of Sight (LoS) nadir angle as seen from the MEO satellite. These linear combinations eliminate the range, clocks, troposphere and the ionosphere delays from the observations. What remains is a combination of code and carrier phase multipath, carrier phase ambiguities, noise, and finally code and carrier antenna effects. The carrier multipath, noise, and antenna effects are much smaller than the corresponding effects for code, thus the carrier multipath and noise can be disregarded. The presence of the carrier phase ambiguities means that it is not possible to perform an absolute evaluation of the antenna effects. To eliminate this ambiguity the multipath observables have been corrected at each epoch with the average multipath observable over the arc. From input measurements up to output calibrated biases, the internal behaviour of the system function and the associated data flow are analysed and the overall system performance evaluated in term of accuracy achieved in the knowledge of the SISBAF, and estimation accuracy of the TEC. It is also evaluated which should be the impact on the Single Frequency User Equivalent Range Error (SF-UERE). The SISBAF estimation and prediction function is in charge of processing the measurement data coming from the LEO S/C, estimating the SISBAF values, and predicting their evolution in the near future. The pre-processing is critical for obtaining precise results, in particular for the LEO data. Several cycle slip detection strategies are evaluated, as well as different smoothing algorithms including divergence-free smoothing. Both original pseudorange and geometry-free combination are used in the smoothing process, and different techniques for multipath mitigation and noise reduction are considered, since these effects are significant error sources for LEO observations. The estimation process is made up of two components: the measurement generation, which prepares the measurements for the estimation process; and the filtering, which applies an EKF (Extended Kalman Filter) to obtain the SISBAF parameters. This is a sequential scheme that produces an estimate of the state at each observation epoch as new measurements become available. CALIBRATION RESULTS In the next months an extensive validation campaign using real data from selected LEO missions will be executed with the objective to demonstrate the validity of preliminary results presented in [1], which were obtained with a simplified Matlab prototype. It is expected to demonstrate that accuracy at centimetric level can be achieved, and the estimation converges to steady state solution after very short transient. CONCLUSIONS Detailed description of SISBAF calibration function and associated data flow will be provided in the paper, along with the results of an extensive validation campaign that will be performed during the next months. It is expected to obtain a very accurate bias calibration. REFERENCES [1] P. D’Angelo, J. Pulido, A. Fernández, J. Bandeiras, P. Vieira, P. Silva, M. Bento, F. Amarillo, “GNSS-BICS – GNSS Bias Calibration System”, Proceedings of ION GNSS 2012 conference, Nashville, Tennessee, September 2012.
Published in:	Proceedings of the 26th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2013) September 16 - 20, 2013 Nashville Convention Center, Nashville, Tennessee Nashville, TN
Pages:	2440 - 2446
Cite this article:	D’Angelo, P., Pulido, J., Rincón, D., Silva, P., Vieira, P., Amarillo, F., "GNSS Bias Calibration Process and Results," Proceedings of the 26th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2013), Nashville, TN, September 2013, pp. 2440-2446.
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In