Theory And Method Of GNSS Ambiguity Resolution For Smartphones

Global Navigation Satellite Systems(GNSS)and their combined systems with other sensors can provide centimeter-level or even higher precision positions and play an important role in applications such as precision navigation,deformation monitoring,and earthquake early warning.However,such a level of accuracy is obtained by performing complex processing on expensive survey-grade receivers and antennas.With the development of microelectronics technology,the performance of miniaturized,low-cost GNSS chips and inertial sensors has gradually improved,giving a glimpse of the potential for high-precision positioning using low-cost devices.Smartphones are integrated with consumer-grade GNSS chips and inertial sensors,providing an effective research platform for tapping the potential of miniaturized,low-cost sensors for high-precision positioning.Smartphones account for almost 80%of the installed base of GNSS devices worldwide,and the number of smartphones is increasing,which contains an unlimited commercial value and vast potential market.Meanwhile,the mass application market and new application industries have an increasing demand for mass high-precision positioning,and they are eagerly expecting the technical support of decimeter-level or even centimeter-level GNSS positioning.However,the GNSS observations of smartphones suffer from severe multipath errors and phase biases,resulting in unresolved carrier phase ambiguity.Therefore,the GNSS positioning precision of smartphones remains trapped in the meter-level,which cannot meet the needs of the above-mentioned mass-market applications and the development of emerging application industries.Therefore,this dissertation uses smartphones as a research platform to verify the feasibility of centimeter-level high-precision positioning using consumer-grade GNSS chips,antennas,and inertial sensors.This dissertation describes an in-depth study on the key technologies of GNSS ambiguity resolution and other high-precision positioning for smartphones.The error characteristics of smartphone GNSS were analyzed,and the carrier phase bias estimation method and multipath mitigation method were proposed,and the purpose of centimeter-level high-precision positioning of smartphones was achieved.Based on this,we further explored methods such as synchronous integration of smartphone GNSS with the accelerometer to pursue higher precision and higher resolution positioning and velocimetry results.Therefore,these works can provide theoretical methods and technical support for high-precision positioning using miniaturized,low-cost GNSS and inertial sensors.The primary contents and contributions of this dissertation can be summarized as the following aspects.1.We systematically analyzed the quality and error characteristics of recent smartphone multi-GNSS observations.We found that the carrier-to-noise density ratio(C/N₀)of smartphones are about 10 d B-Hz lower than those of geodetic receivers and are characterized by rapid variations and low values at high elevations.In terms of measurement noise,the results demonstrated the pseudorange noise and carrier phase noises are 2-10 times and 2-5 times larger than that of the geodetic receivers respectively,and discernible differences among devices,constellations and frequencies.In addition,duty-cycle,anomalous“jagged”distribution phase error,clock misalignment,and inconsistent pseudorange and phase clocks in smartphone GNSS observations have also been identified.Based on different methods,we provide theoretical parameters for the noise versus C/N₀models of the GNSS chipset for different smartphones.2.We proposed a double-difference carrier phase bias and inter-frequency phase bias(IFB)rate extraction method based on the zero/short baseline.By using this method,we found that the smartphones including Nexus 9,Samsung Galaxy S8,and Huawei Honor v8 all have unaligned chipset initial phase biases.Moreover,for Xiaomi 8 smartphone,the phase biases of BDS B1I,Galileo E5a,QZSS L1 frequencies,and the phase IFB rate inconsistent issues of the GLONASS G1 frequencies were found.For this problem,we proposed a gain filtering-based online phase biases correction method,effectively eliminates the influence of phase biases and overcomes the above-mentioned IFB rate inconsistency problem.Based on this method,we implemented dual-frequency and full-constellation GNSS ambiguity resolution on the Xiaomi8 smartphone connected with an external survey-grade antenna.Compared with dual-frequency GPS and full-constellation GNSS without phase biases correction,the ambiguity-fixing rate is improved by 30.4%and 99.2%respectively,and the positioning accuracy is improved by 93.4%and 41.8%respectively.3.For the severe multipath errors on GNSS observations when smartphones using their embedded GNSS antenna,we proposed a multipath mitigation method based on the stochastic model of double-difference code-minus-carrier(CMC)combined observations.The results show that the method can accurately reflect the influence of measurement noise and multipath delay in real-time,and adjust the weight of observation by the random model.Based on this method,we implemented dual-frequency and multi-system GNSS ambiguity resolution on the smartphone using its embedded GNSS antenna.Compared with the classical elevation-based and C/N0-based random models,the ambiguity-fixing rate is improved by about 50%and 30%respectively,and the positioning accuracy is improved to the centimeter level.On this basis,we further explored a multipath mitigation method combining stochastic model compensation and functional model correction.This method inherits the advantages of the CMC stochastic model compensation method,and uses multipath time correlation characteristics to correct multipath errors in real-time,which further improves the performance of ambiguity resolution of smartphones in static positioning.4.We developed an application study of high-precision broadband positioning based on consumer-grade GNSS chips and accelerometers using smartphones as the research platform.A method to synchronize and integrate Android GNSS with accelerometer data is proposed to capture broadband vibration displacements at sub-centimeter resolution.Based on this synchronous integration strategy,a single-receiver GNSS/acceleration tight integrated positioning method based on inter-satellite difference is further proposed to solve the problems of GNSS phase clock instability,inconsistent pseudorange and phase clocks,and the generation time misalignment of smartphones.Experimental results show that smartphones using this method can capture broadband vibrations at centimeter resolution without any augmentation correction input.