| The location service provided by smartphones plays an important role in people’s daily life.With the development and progress of society,users have higher requirements for the accuracy of location services.However,the built-in navigation and positioning module of smartphones only supports output location,speed and other information.In the absence of original data,developers can not do more in-depth research and analysis.In 2016,Google announced that the Android 7.0 system supports the output of information related to the generation of pseudorange,phase and other observations,which provides convenience for relevant scholars to study positioning algorithms based on GNSS observation data,and also brings possibilities for improving the accuracy of smartphone location services.However,smartphones are equipped with poor-performance GNSS antennas and low-cost navigation and positioning chips.The data quality of GNSS observations is bound to be affected,and the positioning accuracy will also be affected.Therefore,based on the smartphone GNSS observation data,this thesis focuses on the positioning performance of smartphone GNSS algorithm and GNSS/PDR fusion algorithm,and develops a real-time fusion positioning software.The research contents and conclusions can be summarized as follows:(1)The quality of smartphone GNSS data is evaluated.The quality of GNSS data is evaluated from the aspects of satellite visibility,signal strength,DOP,data integrity and observation noise.The L5/E5 a frequency observation data is analyzed emphatically.The analysis results show that the L5/E5 a frequency observation value of smartphone has the problems of discontinuous signal tracking and less available satellites.The average SNR of the smartphone L5/E5 a frequency is less than that of the L1/E1 frequency,and the correlation between the second frequency SNR and the elevation angle is higher than that of the first frequency.The pseudorange observation accuracy of smartphone L5/E5 a frequency is better than that of L1/E1 frequency.The pseudorange noise of GPS system is greater than that of BDS and Galileo system,and the phase noise of Galileo system is greater than that of BDS and GPS system.(2)The positioning performance of smartphone pseudorange single point positioning algorithm is analyzed.The positioning performance is analyzed from the aspects of the number of satellite systems participating in the solution,the pseudorange smoothing method,the stochastic model,and the number of signal frequencies.The positioning performance of the inter-satellite differential SPP model and the standard SPP model is compared and analyzed.The experimental results show that the pseudorange after doppler or phase smoothing can improve the positioning accuracy,and the doppler smoothing method performs better.The SNR model is more suitable for processing smartphone data,and the positioning effect is better than other models.The positioning accuracy of the dual-frequency algorithm in all directions is higher than that of the single-frequency algorithm,and the elevation direction is significantly improved.The plane positioning accuracy of the inter-satellite difference model and the standard model is similar,but the former has better positioning performance in the elevation direction.(3)The processing strategy of smartphone precise point positioning algorithm is optimized.Based on the single-frequency PPP model with ionospheric constraints and the dual-clock offsets PPP model with ionospheric constraints,the data processing method is optimized,and an uncombined dual-clock offsets PPP model for unified processing of single-frequency and dual-frequency observation data is formed to make full use of GNSS observation information.The positioning performance of different models is compared and analyzed by experiments.The results show that compared with other models,the single and dual-frequency model has better positioning effect,the plane and elevation positioning accuracy are improved,and the elevation direction is improved significantly.In the static mode,the plane accuracy of the model is 0.466 m,and the elevation accuracy is 0.351 m.In the dynamic mode,the plane accuracy of the model is 1.214 m and the elevation accuracy is 1.453 m.(4)The positioning performance of smartphone differential positioning algorithm is analyzed.According to the original observation equations of pseudorange and phase,the mathematical model of differential positioning is derived,and the positioning performance of pseudorange difference and carrier phase difference is analyzed through experiments.The experimental results show that the positioning effect of the pseudorange difference algorithm is better than that of the single-frequency SPP algorithm in both static and dynamic experiments.The positioning accuracy of the phase difference algorithm is better than that of the single and dual-frequency model in the static experiment.In the dynamic experiment,the elevation accuracy is similar to the single and dual-frequency model,and the plane accuracy is improved by 7 %.(5)The effectiveness of the fusion positioning algorithm in different environments is verified and a real-time fusion positioning software is developed based on the algorithm.Firstly,the algorithms used in step frequency detection,step length estimation and heading estimation model are determined by experiments.Then,the positioning performance of GNSS/PDR fusion algorithm in open environment,semi-open and semi-occlusion environment,challenge environment is analyzed.The experimental results show that compared with the smartphone GNSS algorithm,PDR algorithm and the chip solution inside the smartphone,the fusion positioning algorithm has higher accuracy and better stability,which effectively improves the unstable positioning effect of the smartphone using a single algorithm.At the same time,based on the Android smartphone,a software that can perform GNSS/PDR fusion positioning in real time is developed.The software has real-time positioning,data analysis,data display,positioning parameter setting,map visualization and other functions,and the availability of the software is verified by experiments. |
PDF Full Text Request