| Abstract |
The last decade has seen substantial advancements in Internet of Things (IoT)-based transportation and smart city networks, fueling the growth of Global Navigation Satellite System (GNSS) industries and GNSS-enabled smartphones that deliver real-time, precise location-based services for mass-market applications. However, achieving decimeter-level smartphone positioning with GNSS processing techniques, such as precise point positioning (PPP) with real-time corrections in urban environments remains challenging due to the noisy and unstable nature of smartphone GNSS measurements. Key issues include low signal strength, high multipath effects, frequent cycle slips, and phase discontinuities, all of which degrade PPP accuracy and extend convergence times. To address these challenges, this study introduces a two-step clock bias preprocessing method to reduce Galileo High Accuracy Service outliers and biases. Additionally, an innovative iterative PPP algorithm integrated with a moving window approach is proposed to mitigate cycle slip false alarms and preserve ambiguity estimation continuity under difficult GNSS signal reception conditions. Validated through extensive vehicle experiments across eight datasets in diverse multipath environments, the proposed method demonstrates significant positioning accuracy improvements with four-constellation support. Results show a 95th percentile error and overall rms of 1.8 and 1.2 m, respectively, in horizontal positioning, with submeter lateral rms (0.8 m) and 99% lane-determination success rate in realistic driving scenarios. These findings indicate the potential of smartphone-based real-time PPP in enabling lane-level navigation, paving the way for next-generation IoT-integrated location services. © 2014 IEEE. |
