GNSS Real-Time Precise Point Positioning Method For Smart Phones And Analysis Of Accuracy Influencing Factors

Using smart phones to achieve precise,continuous and reliable real-time positioning result is a notable subject.From 2016,Google announced to output the raw Global Navigation Satellite System(GNSS)measurement data on the intelligent terminal of Android Nougat 7.0 and above operating system;and with the development of IGS-RTPP(IGS Real Time Pilot Project),International GNSS Service(IGS)analysis centers began to provide users with satellite orbit and satellite clock in the form of State Space Representation(SSR)in real time.The real-time precise point positioning based on smart phone becomes possible.However,the RTS(Real Time Service)products provided by some IGS are deficient,some satellite systems and satellites are lack of products.And it is easy to cause data delay or loss in case of network congestion or interruption.Therefore,the RTS products received by users are often incomplete,which brings challenges to continuous precise real-time positioning.Aiming at the above problems,this paper proposes a real-time smart phone precise point positioning method considering the absence or incompleteness of RTS products,and improves the positioning strategy by the refinement of observation values random model and the optimization of a priori ionospheric noise setting.The main work contents are as follows:(1)Aiming at the problem that RTS products are incomplete or some satellites are missing,considering that there is more gross error in smart phone observation data in urban environment,in order to make full use of all available observation data,a realtime precise positioning method of smart phone using RTS products and broadcast ephemeris is proposed.The state equation is constructed by the time differenced carrier phase method,and the observation equation is established based on the undifferenced uncombined model.Combined with the constraints of a priori ionospheric model,the kalman filter for real-time precise positioning of smart phone is designed and verified by static and dynamic experiments.(2)Aiming at the problem that RTS satellite orbit and satellite clock are incomplete in real-time positioning,based on the accuracy of different real-time satellite orbit and clock,propose a refinement strategy of observation value random model.By properly setting the noise of satellite observations,improve the continuity and positioning performance of smartphone navigation.The static test result shows that by using the refinement strategy,the E/N/U direction accuracy can be improved by 17%～54%,and the convergence can be effectively accelerated.(3)Aiming at the lack of ionospheric products in some RTS products,by analyzing the correction accuracy of klobuchar model and real-time spherical harmonic model,a stochastic model optimization strategy with ionospheric constraints is proposed.According to the accuracy of the ionospheric model used,an appropriate priori ionospheric noise is set to improve the convergence speed and positioning performance.The experiment shows that when the real-time spherical harmonic model is used,the priori ionospheric noise is set to 0.5m;when Klobuchar model is used,it is set to 1.5m.Compared with the unconstrained method,the positioning accuracy and convergence speed of smart phone are significantly improved.(4)To verify the positioning performance of the smart phone real-time precise positioning method using different products,static and dynamic tests in suburban areas are carried out with Xiaomi Mi 8 and Huawei Mate 40 Pro.The results show that using mixed products,the static positioning accuracy of Mi 8 can reach 0.6m,the dynamic positioning accuracy can reach 1.6m,while the positioning accuracy of Mate 40 Pro is slightly worse,which is equivalent to that of RTS products only;If only use broadcast products.The dynamic positioning accuracy of Mi 8 can reach 1.8m,and that of Mate40 Pro can reach 5.5m,which is slightly lower than the mixed use scheme.This is because the pseudo range noise of smart phones is large,up to 1～5m,which is larger than the accuracy of most RTS products.Making full use of all available observation data will help to improve the positioning performance of smart phones..