Title: Quality Analysis for Satellite Bias Estimation and GNSS PPP Ambiguity Resolution
Author(s): Shuyang Cheng
Published in: Proceedings of the 30th International Technical Meeting of The Satellite Division of the Institute of Navigation (ION GNSS+ 2017)
September 25 - 29, 2017
Oregon Convention Center
Portland, Oregon
Pages: 2219 - 2234
Cite this article: Cheng, Shuyang, "Quality Analysis for Satellite Bias Estimation and GNSS PPP Ambiguity Resolution," Proceedings of the 30th International Technical Meeting of The Satellite Division of the Institute of Navigation (ION GNSS+ 2017), Portland, Oregon, September 2017, pp. 2219-2234.
Full Paper: ION Members/Non-Members: 1 Download Credit
Sign In
Abstract: *ION GNSS+ 2017 Student Paper Award Winner* Ambiguity resolution enabled PPP, called PPP-AR, has attracted great attention from the GNSS community over the years. PPP-AR aims to separate the satellite biases from integer ambiguity or compensate these biases in float ambiguities and then improve the positioning accuracy with short convergence time. Three representative PPP-AR methods are well-known up to now: fractional cycle bias (FCB) method, integer recovery clock (IRC) method and decoupled satellite clock (DSC) method. In all these methods, satellite products are generated from a global or regional reference network and then broadcast to users for ambiguity fixing. PPP-AR appreciably improves the positioning accuracy and reduce the converge time. However, highaccuracy satellite bias products are the prerequisite of reliable GNSS PPP ambiguity resolution, and thus quality control for the estimation of satellite bias products is crucial to PPP-AR. In addition, the trade-off between the reliability and computational efficiency in the satellite products generation needs to be taken into account. In this paper, we propose an estimation and quality control strategy for the FCB and IRC estimation based on IGS clock products. A dense and a sparse global network are used to analyze the impact of reference station quantity and distribution on the estimated FCB and IRC products A joint-processing model for GPS/GLONASS/BDS triple-system combined PPP with GPS and BDS ambiguity fixing based on raw observations is also elaborated and a partial ambiguity resolution (PAR) strategy is implemented in PPP. Datasets from IGS MGEX network and CORSnet-NSW in Australia are used to validate the proposed methodology. The experimental results indicate the good reliability in the Least squares-based FCB and IRC estimation. A global sparse network which consists of 72 stations is suggested to be involved in FCB and IRC products generation to reduce computation burden. With these satellite biases products, the positioning biases in the ambiguity-fixed solution for triple-system PPP-AR can be reduced to 0.7 cm, 0.45 cm and 2 cm in east, north and up direction, in comparison with 1.2 cm, 0.5 cm, and 2.4 cm for GPS-only PPP-AR, respectively, and the time-to-first-fix (TTFF) can be shortened by about 10 minutes. PAR can further improve the PPP-AR performance, in terms of TTFF and fixing rate. Moreover, both FCB based PPP-AR and IRC based PPP-AR can achieve similar performance.