An Improved Model For Precise Point Positioning With Modernized Global Positioning System

Abstract

Recent developments in GPS positioning show that a user with a standalone GPS receiver can obtain positioning accuracy comparable to that of carrier-phase-based differential positioning. Such a technique is commonly known as precise point positioning (PPP). A significant challenge of PPP, however, is that it typically requires a minimum of 30 minutes to achieve centimeter- to decimeter-level accuracy. This relatively long convergence time is the result of un-modeled GPS residual errors. This thesis addresses error mitigation techniques to achieve near real-time PPP. To explore the full advantage of the modernized GPS L2C signal, it is essential to determine its stochastic characteristics and code bias. GPS measurements were collected in order to study the stochastic characteristics of the modernized GPS L2C signal. As a byproduct, the stochastic characteristics of the legacy GPS signals, namely C/A and P2 codes, were also determined and then used to verify the developed stochastic model of the modernized signal. The differential code biases between P2 and C2, DCB P2-C2, were also estimated using the Bernese GPS software. A major residual error component, which affects the convergence of PPP solution, is the higherorder ionospheric delay. We rigorously modeled the second-order ionospheric delay, which represents the bulk of higher-order ionospheric delay, for our PPP model. First, we investigated the effect of second-order ionospheric delay on GPS satellite orbit and clock corrections. Second, we used the estimated satellite orbit and clock corrections to process the GPS data from several IGS stations after correcting the data for the effect of second-order ionospheric delay. The results demonstrated an improvement of up to 25% in the precision of the estimated coordinates with the second-order ionospheric delay, as well as reduction of the convergence time of the estimated parameters by about 15%, depending on the geographic location and ionospheric and geomagnetic conditions. Between-satellite single-difference PPP algorithms were developed to cancel out the receiver clock error, receiver initial phase bias, and receiver hardware delay. The decoupled clock corrections, provided by NRCan, were also applied to account for the satellite hardware delay and satellite initial phase bias. GPS data collected from several IGS stations were processed using the un-differenced model, un-differenced decoupled clock model, between-satellite singledifference (BSSD) model, and between-satellite single-difference using the decoupled clock (BSSD-DC) model. The results showed that the proposed BSSD model significantly improved the PPP convergence time by 50% and improved the solution precision by more than 60% over the traditional un-differenced PPP model.