Download files
Abstract
In the last two decades Precise Point Positioning (PPP) has become a well-established stand-alone positioning method by means of Global Navigation Satellite Systems (GNSSs) for a wide range of applications. By using code and phase measurements from a single GNSS receiver and precise orbit and clock information derived from global or regional GNSS networks, highly precise positions can be obtained. One critical problem of the PPP technique is that a typical period of about 30 minutes is required to reach decimeter-level under normal conditions.
In order to shorten the convergence time and improve the positioning accuracy, several PPP integer ambiguity resolution methods have been developed in the last decades. The common approach is to provide additional corrections on the existing PPP (clock) products. Although improvement in positioning accuracy is achieved by fixing ambiguities, the initialization time of PPP is not significantly reduced. Rapid convergence and ambiguity resolution in PPP is still a challenge. The key to instantaneous ambiguity resolution in relative positioning for short baselines lies in the a priori knowledge of the ionosphere. Therefore, the primary objective of this thesis is to investigate the introduction of global and regional ionospheric models as external constraints for enhancing PPP ambiguity resolution. The main work and contributions of this thesis are specified as follows:
a. Mathematical modeling aspects for PPP are investigated. The procedures of estimating FCBs for integer ambiguity resolution in PPP based on standard ionosphere-free model and uncombined model are derived. The compatibility of FCBs estimated from both models are validated by comparing their wide-lane and narrow-lane float ambiguities as well as estimated FCBs using real data sets.
b. The equivalence of three extended PPP integer ambiguity resolution models with capabilities of constraining external ionosphere information is derived. The method equivalence is demonstrated in three aspects: the ionospheric parameter, integer property recovery and the system redundancies. It is shown that all three models permit strengthening solutions by constraining ionospheric parameter from global and regional ionospheric models. The positioning results indicate that PPP can be further improved if external ionospheric information is available.
c. Accuracies of regional atmospheric corrections for PPP ambiguity resolution are assessed. The focus is on the achievable accuracies of interpolated tropospheric and ionospheric delays derived from a typical regional network from ambiguity-fixed PPP solutions. The results indicate that centimeter level accuracies can be obtained for both tropospheric and ionospheric corrections and fast ambiguity resolution can be achieved after applying the regional atmospheric corrections.
d. Ambiguity-fixing approaches based on uncombined PPP model are investigated, the one-step approach, in which L1 and L2 ambiguities are simultaneously fixed, and the two-step approach which involves sequentially fixing wide-lane and N1 ambiguities such that the fixed wide-lane ambiguities are applied as constraints to update remaining unknown parameters. Experiment results demonstrate that ambiguity-fixing time can be reduced using the two-step approach as compared to the one-step approach.
e. Software system is developed to estimate satellite FCBs from global and regional GNSS network using both ionosphere-free and uncombined model. A PPP software package is developed to validate the contribution of global and regional ionospheric information on PPP ambiguity resolution.