Research On GNSS Network RTK Positioning Principle And Algorithm

With the rapid development of differential positioning technology of Global Navigation Satellite System(GNSS),the current conventional RTK technology of GNSS has some shortcomings in both the scope of action and the quality control and guarantee of positioning solution.The Network RTK technology based on the continuous operation reference station(CORS)system emerges as the times require and becomes a research hotspot in the new century.Therefore,it is of great practical significance to study the RTK Positioning Principle and algorithm of GNSS network.The main research content and The innovations are as follows:(1)The GNSS network RTK positioning mathematical model is analyzed and designed in detail,including the time-space unified conversion calculation of each GNSS system,and the atmospheric delay error that cannot be eliminated under the network RTK double-difference positioning model is analyzed and corrected.(2)Due to the interference of atmospheric factors,and when the new satellite rises above the cut-off altitude,it needs a long initial convergence time,an improved method for fast resolution of ambiguity between network RTK reference stations of GNSS is proposed.Firstly,according to the prior information,the pseudo observation model of Gauss Markov is designed by using the ionosphere weighting strategy,so as to assist the fast calculation of baseline ambiguity.The experimental results show that,compared with the solution without ionosphere weighting model,the convergence time of both the initialization time and the rising time of the new satellite after weighting is reduced.On the basis of weighting,the extended kalman filter(EKF)technology is used to estimate the real value of the ambiguity,and then the improved partial ambiguity(PAR)solution is used to select a fixed subset of ambiguities.Finally,the LAMBDA and RATIO detection methods are used to solve the fixed parameters of ambiguities degree.And the data experiments prove that the method not only shortens the initial convergence time of the long-term baseline ambiguity solution in the GNSS network RTK,but also overcomes the problem of fixed success rate of ambiguity during satellite ascent and descent.At the same time,compared with FAR and conventional PAR methods,it not only improves the ambiguity resolution efficiency and success rate of GNSS network RTK reference stations,but also makes the baseline positioning accuracy reach the centimeter level.(3)In view of the compatibility and interoperability of GNSS systems in the future,the solution efficiency of multi-frequency,multi-mode and high-dimensional ambiguity in conventional methods is low,based on the lattice theory,the algorithm of GNSS ambiguity solution is improved,and the closest lattice point(CLP)algorithm is proposed to search the ambiguity shaping value.In this method,firstly,the ambiguity search is transformed into the nearest grid point search problem of the known grid points in the lattice.Then,according to the lattice base specification,the lattice base vectors with the minimum possible length and orthogonal to each other are obtained.Finally,CLP algorithm is used to search the optimal ambiguity parameters.Based on lattice theory,the proposed CLP search algorithm is theoretically more efficient and reliable than the classical lambda /MLAMBDA algorithm,and the search time of each parameter is stable at 0.01 seconds,even in high-dimensional case,the search efficiency of CLP algorithm is still stable and reliable.The ambiguity solution method based on lattice theory breaks through the conventional solution idea,which brings new opportunities for the high-precision solution of multi frequency and multi-mode system in the future.