Enhancing smartphone precise point positioning to sub-meter accuracy in suburban environments: a new stochastic model and outlier diagnosis

Jiahuan Hu,Pan Li,Sunil Bisnath
DOI: https://doi.org/10.1007/s10291-024-01651-5
2024-04-26
GPS Solutions
Abstract:Smartphone global navigation satellite system measurements in suburban environments suffer from severe multipath and noise, posing a challenge in dealing with the potential outliers. Using the range errors derived from smartphones, the characteristics of range errors can be further utilized to optimize the stochastic model and improve outlier diagnosis in user environments. Compared to existing algorithms, a new adaptive phase/code precision ratio scheme is proposed. Furthermore, a scale factor is utilized to deal with the over-optimization issue in the fitted weighting scheme functions, and adaptive a posteriori residual rejection method which considers the number of rejected measurements is introduced. The positioning improvements brought by the proposed algorithm are validated step by step. The proposed adaptive phase/code precision ratio can reduce the 68th and 95th horizontal positioning errors by 4 cm and 3 cm, respectively, compared to a ratio of 100. The percentage of horizontal positioning errors within 1 m can be improved by 7% (from 50 to 58%). Compared to the conventional carrier-to-noise ( C / N 0 )-based weighting scheme, the proposed weighting scheme can reduce the 68th and 95th percentile horizontal positioning errors by 14 cm and 15 cm, representing improvements of 11% and 8%, respectively. Applying the new residual rejection method, the percentage of positioning errors within 1 m improves from 42 to 54% compared to C / N 0 -based weighting scheme. The results demonstrate great potential for achieving overall sub-meter level horizontal positioning performance with smartphones in suburban environments.
remote sensing
What problem does this paper attempt to address?