A robust outlier detection method is proposed for effective detection of smartphone Global Navigation Satellite System (GNSS) measurement outliers frequently caused by multipath and non-line-of-sight (NLOS) effects. The new method has been developed through a novel combination of a binary-tree-based solution separation (SS) test (SS-test) and the deep-learning techniques, which leverages the statistical testing’s outlier detection efficiency and the deep-learning’s proficiency in modeling complex relationships. First, the binary-tree-based SS-test identifies specific subsets of outlier-free measurements. Second, the network predicts three-dimensional (3D) positioning errors through those subsets. Third, we obtain an accurate positioning solution from the measurement subset with the smallest predicted errors. To validate the proposed approach, a vehicle-based field test was conducted with a smartphone. The testing results indicate that the proposed method has reduced the rate of large positioning errors by 79% and improved the positioning accuracy by 41% compared to the conventional methods without robust statistical testing.

Ionospheric delay is a dominant error source in global navigation satellite systems (GNSSs). Conventional ionospheric estimation relies on dense networks of expensive geodetic receivers, limiting accessibility and coverage. With the widespread availability of multi-frequency, multi-constellation smartphones capable of carrier-phase tracking, this study investigates smartphone-based ionospheric estimation. Using a single-reference Precise Point Positioning Real-Time Kinematic (PPP-RTK) framework, ionospheric delays are estimated from smartphone data and evaluated using real-time correction products from BeiDou PPP-B2b and Centre National d’Études Spatiales (CNES). Quality control is performed via solution separation testing with time-differenced carrier phase and time-differenced pseudorange. Field experiments with two Google smartphones and a geodetic receiver demonstrate that the estimated slant ionospheric accuracy is comparable to geodetic receivers within the meter level under both static and kinematic scenarios. Additionally, the horizontal positioning performance demonstrates that the positioning performance of the user smartphone with ionospheric corrections broadcast from the base smartphone is significantly improved, with 74.7% and 54.9% for CNES and PPP-B2b products compared with the conventional PPP solution. Furthermore, a comparison between ionospheric corrections generated from the smartphone and those obtained from the geodetic receiver reveals that the positioning performance of the user smartphone becomes comparable after convergence.

This study proposes a methodology to achieve accurate vehicle maneuver detection and risk analysis, innovatively leveraging the smartphone time-differenced carrier-phase (TDCP) measurements and the factor graph optimization (FGO) method for accurate vehicle motion estimation. With the centimeter-level precision of TDCP, the method achieves accurate vehicle displacement measurements, while the FGO framework enhances the estimation of vehicle speed, acceleration, pitch and heading angles, as well as angular velocity. Extensive experiments involving multiple vehicles and drivers validate the method’s effectiveness in vehicle motion estimation, maneuver identification, and risk classification. The results demonstrate a reduction in speed and angular velocity estimation errors by up to 47.28% and 66.95%, respectively, compared to traditional GNSS solutions. Furthermore, classification accuracy for high-risk driving behaviors improves by 27.37%, making the method suitable for applications such as usage-based insurance (UBI), fleet management, and autonomous driving.

High accuracy smartphone positioning is a research topic under extensive investigation driven by the availability of raw GNSS (GPS, Galileo, GLONASS, Beidou) carrier phase observations to Android smartphones users and great value to location-based services (van Diggelen et al, 2018; Zangenehnejad & Gao, 2021). Location-based services are to find out the geographical location of the mobile user and then provide services based on the given location information and they can touch almost every aspect of our daily life from health care, transportation, emergency and rescue services, public safety, to homeland security. High accuracy spatial location information is a vital component in developing location-based service to mobile devices, including government mandates that cellular operators must provide location of emergency callers within a certain degree of accuracy. Today, nearly every new smartphone contains a GNSS chip. While the accuracy of standard GNSS positioning with smartphones usually in the range of 5 to tens of meters, improved positioning accuracy is expected to enable many new location-based applications. This is because high accuracy will make it possible to deliver personalized services of great value to mass market applications. Better positioning accuracy, for instance, can provide new experiences in lane-level navigation and augmented reality with sensor fusion. Professional apps such as surveying could also take advantage of high accuracy positioning using much cheaper multi-constellation and multi-frequency smartphones.

The widespread adoption of smartphones has spurred advancements in navigation technology. Traditional positioning methods, such as precise point positioning (PPP) and real-time kinematic (RTK), face limitations including long convergence times and high data rate requirements, making them unsuitable for smartphone applications. The PPP-RTK approach offers faster ambiguity resolution by leveraging precise atmospheric corrections, which is expensive to install and maintain from Continuously Operating Reference Stations (CORS) or multiple reference stations. Although PPP-RTK from single reference station is developed, satellite orbital/clock and code bias corrections from ground-based CORS stations and data transmissions are still required. While using the satellite-based Galileo high accuracy service (HAS) can achieve high-accuracy positioning for electronic devices including smartphones, few studies have studied HAS corrections to assess its effectiveness. In this study, we propose a smartphone precise positioning method from PPP-RTK with HAS and single reference station. Moreover, due to the high noise level of smartphone observations, a timedifferenced carrier-phase (TDCP) model coupled with solution separation (SS) testing is used to detect cycle slips. To validate the proposed method, experiments are carried out in static and kinematic modes. The results show the proposed method can achieve convergence in both static and kinematic condition around 2 minutes with ambiguity resolution, achieving an accuracy of 0.411 m horizontally and 0.754 m in three-dimension in kinematic mode. It suggests that utilizing current HAS products with GPS and Galileo smartphone PPP-RTK methodology can enable decimeter-level positioning precision.

The ionospheric delay is a major error source in the Global Navigation Satellite System (GNSS) positioning, and its accurate estimation is essential for Precise Point Positioning Real-Time Kinematic (PPP-RTK). Traditionally, ionospheric delay estimation relies on a network of permanently deployed high-end geodetic GNSS receivers, which are costly and thus inaccessible for most consumer applications. Moreover, this approach is limited by the sparse spatial distribution and low temporal resolution of the network, leading to significant estimation errors in uncovered environments.On the other hand, due to the global density and accessibility of low-cost GNSS receivers, such as smartphones, there is a strong demand to develop new methods for precise ionospheric delay estimation using them. Moreover, multi-frequency and multi-constellation GNSS chipsets are now embedded in smartphones including carrier phase observations essential for precise positioning. These advances support the investigation and development of new methods to enable precise real-time GNSS positioning even using smartphones. However, few studies have focused on the application and evaluation of such methods for PPP-RTK positioning.Therefore, this study aims to develop methods to estimate ionospheric effects using low-cost GNSS receivers and demonstrate that it can provide reliable ionospheric corrections. Additionally, we evaluated ionospheric corrections using two real-time satellite orbit, clock, and code bias products, namely the satellite-based BeiDou PPP-B2b and ground-based Centre National d’Etudes Spatiales (CNES). First, the ionospheric delay estimates generated by a single reference smartphone with uncombined PPP and quality control measures based on solution separation testing is evaluated using of the real-time satellite orbit, clock, and code bias products from BeiDou PPP-B2b and CNES, respectively. Next, the generated ionospheric delay from two correction models is compared to that produced by a high-end geodetic receiver. Finally, the generated ionospheric corrections are applied to single-station-based PPP-RTK to assess its positioning performance under kinematic conditions. A field test was conducted using two Google smartphones on April 7, 2024, in Calgary. We expect to achieve decimeter-level slant ionospheric corrections accuracy compared to geodetic receiver with the two correction models used. Additionally, the positioning accuracy is expected to approach that of PPP-RTK results using geodetic receivers as base stations, significantly outperforming float PPP.