Research On The Methods For Multi-GNSS Products Combination And Its Application In IGMAS Activities

In the last tens of years,Global Navigation Satellite System(GNSS)has achieved great progress and promotes the geodynamics and atmosphere environmental science with the technique related products(e.g.tracking station coordinates,satellite orbit and clock,Earth rotation parameters(ERP)).The precise products are also applied widely in the engineering and public life.Meanwhile,the GNSS data processing theory and method,the rapid solution for large network,the understanding and modeling of observation errors have to be improved/solved with the rapid development of GNSS(e.g.the emerging of multi-mode and multi-frequency observations,the expansion of global tracking network,the differences in satellite geometrical and optical property,the attitude mode and stability of satellite frequency standard between different satellite systems).Undoubtly,the diversification of GNSS data processing strategies,models and softwares adopted by global agencies,leads to different performance in the products delay,precision and datum.It is of great importance for Earth science and engineering application to obtain combined solutions with high precision,reliable and consistency taking advantage of multianalysis centers products.1)The terrestrial reference frame can be achieved with the long-term station coordinates stored in Software/technique Independent Exchange files,and the results of crustal movement and loading deformation induced by mass redistribution on the Earth surface using GNSS technique are meaningful only under the unified and stable reference frame.2)The clear error characteristics in GNSS products can be found by comparison between the solutions from combination and single analysis center,which can promote the improvement of data processing strategies and models in return.3)It is necessary to monitor and assess the operating state and service performance for GNSS,the reliable assessment results of signal in space and navigation can be obtained using high-quality GNSS combined products.Currently,multi-GNSS satellite orbit and clock products are provided by different analysis centers,and it is necessary to conduct some researches on the multi-GNSS products combination and provide reliable combined solutions for multi-GNSS users.However,GNSS products combination is a whole project,which consists of multiple kinds of products,instead of a single kind of product.So the consistency among station coordinates,satellite orbit and clock,Earth Rotation Parameters must be taken into consideration for the products combination,or the inconsistency combined solutions would result inconsistent precise point positioning using combined satellite orbit and clock products.Moreover,the unmodeled phase center offset or solar radiation pressure,the different strategy on the inter-frequency bias and datum constraints,any of these could induce different performance of product precision,systematic error.So it is of great importance to study the model of product combination and methods of parameter estimation.Last but not least,the foundation of GNSS product combination is an unified and long-term stable datum for different products,so it is the first to realize a highprecision reference frame for products combination using station coordinate series derived from multi-analysis centers.This paper aims to solve the problems described above and focuses on the combination of stations coordinates,satellite orbit and clock,ERP products derived from multi-analysis centers.Three main aspects are investigated in this paper.First,the combination of daily station coordinates from analysis centers is studied,and a long-term reference frame for GNSS combined products is realized using twenty-year daily combined solutions.Second,the multiGNSS satellite orbit and clock combination under the unified reference frame is discussed,the robust and stable combined solutions are obtained.Third,the GNSS product combination software is developed and applied in the international GNSS Monitoring and Assessment System(i GMAS).The main contents and results achieved in this paper are as follows.1.The paper first summarized the current states and progress of GNSS product combination,including station coordinate,satellite orbit and clock products.Then the problems in this field are pointed out which would be solved in this paper.The issues of basic theory and methods for GNSS product combination are investigated,special for the robust parameter estimation,free-network adjustment with rank-deficient problems in the normal equation system(NEQS),variance component estimation,the recovery of NEQS,the removal of a priori/residual constraint,the parameter elimination and transformation.2.This paper analyzed the impact of a priori and residual constraint on the coordinate transformations,and the parameter transformation method is adopted to correct the systematic bias in the station coordinate products.Compared the IGS terrestrial reference frame,the realized reference frame in this paper is at the same level,the precision of station position and velocity at the reference epoch is better than 0.5 mm and 0.1 mm/yr,respectively.the seasonal signals in position time series are analyzed and the effects of Helmert transformation parameters and weight matrix on seasonal signals is validated.Meanwhile,the impacts on noise analyses of GNSS position time series caused by seasonal signal,weight matrix,offset,and Helmert transformation parameters are also analyzed.3.The consistency of orientation datum among station coordinate,ERP,satellite orbit and clock for different analysis centers is validated theoretically and confirmed by practical data.The methods of consistency correction between multi-GNSS orbit and station coordinate,as well as the correction between orbit and clock products,are proposed in this paper.The changes of millimeter would be obtained whether or not the consistency correction considered.4.The methods of multi-GNSS orbit and clock products combination are studied in this paper.In order to avoid the large systematic bias among satellites for different analysis centers,the clock alignment method based on single satellite is adopted.The variance component estimation is applied to assign reasonable weights for analysis centers with different precision.Moreover,the equivalent weight functions are designed to resist the outlier in GNSS products and any analysis centers with anomalous performance.Based on the test data,experimental results show that the combined orbit and clock products are more reliable than solutions of any analysis centers when compared to the IGS products.Meanwhile,the orientation datum is also more stable.The result of kinematic precise point positioning with combined products is at least as accurate(if not more so)than the solutions obtained by the individual analysis centers.5.Based on the methods of station coordinate,satellite orbit,ERP and clock products combination adopted by this paper,the software of GNSS product combination is designed,which has been applied in i GMAS successfully and meet products with different delay(e.g.the ultra-rapid,rapid and final products).It is the first time to obtain the multi-GNSS satellite orbit and clock combined solutions under the unified reference frame.Moreover,the GPS combined solutions are at the same level of precision as the products provided by the internal global navigation satellite system service(IGS).6.For the systematic errors in satellite orbit products validated by satellite laser ranging(SLR),this paper purposed a priori model of solar radiation pressure(SRP)for satellite QZS1 during the precise orbit determination.The results show that the augmented SRP model with the new a priori model significantly improved orbit solutions of about 40% by SLR validation and the ?-angle-dependent errors are almost eliminated.Another systematic error correction procedure for i GMAS BDS and Galileo orbits is also proposed with the SLR residuals of history data.Implemented the procedure before combination,the improvements of root mean square error for combined orbits can reach 70% and 30% for BDS geostationary satellite and Galileo satellites,respectively.