Research On GNSS Single Frequency RTK Positioning Performance Of Smart Phone

With the increasing abundance of GNSS navigation and positioning services and applications,users are placing higher demands on the cost and accuracy of GNSS navigation and positioning technology.In recent years,mass-grade smart terminals(smartphones,smartwatches,tablet PCs,etc.)have become an integral part of people's daily life and production.The services and applications of smart device positioning continue to expand,making high-precision positioning technology for smart devices one of the research hotspots in the field of navigation and positioning.In the early days,smartphone terminals were confined to the closed nature of the original data of smart devices and could only output location information,and the research on GNSS navigation and positioning technology for smart devices was slow,and its horizontal accuracy was meter level.In 2016,the GNSS raw observation data interface of Android smart terminals was released,making it possible to achieve high-precision positioning on smart terminals including smartphones.In this paper,we analyze the quality and characteristics of the raw data,optimize the RTK positioning algorithm for Android smartphones by combining the minimum signal-to-noise ratio,Doppler smoothing algorithm and random observation model based on signal-to-noise ratio,and design experiments to test its positioning performance.The main work and contents of the thesis are as follows:(1)For the smartphone raw GNSS data,the smartphone BDS and GPS data quality were analyzed in terms of the number of visible satellites,satellite signal-to-noise ratio,circumferential hopping,data continuity,pseudorange observation noise and carrier phase observation noise.The results show that the overall number of visible satellites of Android smartphones is not inferior to that of the low-cost single-frequency receiver M8 L,in which the number of visible satellites of GPS is comparable to both,and the average number of visible satellites of BDS is slightly more than 3 for smartphones;the signal-to-noise of GPS and BDS satellites of smartphones is comparable to that of M8 L,and there are certain fluctuations,and there is no obvious correlation with the satellite altitude angle;the carrier phase observation value of smartphones has certain The pseudo-range observations of smartphones are single-differenced with the base station data,and then double-differenced with the reference star,and the double-difference residuals are basically in the range of-4?4m,and there is a certain coarse difference;regardless of the weekly jump,the smartphone carrier phase observations are single-differenced with the reference star,and then triple-differenced among the ephemeris,and the phase quadruple-difference basically falls within the range of-0.03?0.03m;the pseudo-range double-difference,phase quadruple-difference and the satellite signal angles are not significantly correlated.The correlation between the phase quadratic difference and the satellite signal-to-noise ratio is obvious.(2)A method is proposed to optimize the RTK processing for Android smartphones,which first eliminates the data with low signal-to-noise ratio and altitude angle,then uses Doppler observations to smooth the pseudorange,and establishes an observation model based on signal-to-noise ratio to assign a priori weights to the observations.Static and dynamic experiments are designed to study the RTK positioning performance of single BDS system and combined GPS/BDS navigation and positioning system.The experimental results show that: in the static experiment,the horizontal accuracy of the combined GPS/BDS system positioning is less than 1cm and the vertical accuracy is less than 3cm,and the effect of the single BDS system is slightly lower than that of the combined system;compared with the F9 P receiver,the convergence time and accuracy required for the static positioning of smartphone GNSS is slightly insufficient,but under good observation conditions,it can meet the centimeter-level RTK positioning requirements;in the dynamic experiment,the 3D distance STD is 3.2cm.