Theory And Method Of Combined GPS/GLONASS Precise Point Positioning

GPS Precise Point Positioning (PPP) is a new satellite positioning technique in which precise satellite orbit and clock data, dual-frequency pseudo-range and carrier phase observables from one receiver are used to determine the position solution. PPP can be conducted globally and ITRF coordinates with centimeter-level accuracy can be achieved directly. In comparison with relative positioning, PPP has a few advantages such as no restriction by the inter-station distance, direct determination of position solution and simple data processing. As a result, PPP is widely applied to the maintenance of high-accuracy reference frame, scientific investigation, high-accuracy kinematic navigation and positioning, orbit determination of low orbit satellites, high-accuracy engineering survey, land resource investigation and so on. However for such a satellite-based navigation and positioning system, the positioning accuracy and the reliability of positioning results are quite dependent on the number of visible satellites. In some situations such as urban canyons, open-pit mines and mountainous areas, the number of visible satellites is insufficient to provide a position solution. A good strategy is to integrate GPS and GLONASS.Since the availability of GLONASS precise ephemeris and clock products, the combined GPS/GLONASS precise point positioning has become possible. In this thesis, the main research contents are as follows. Firstly GPS is systematically compared to GLONASS. Secondly error sources and corresponding handling methods are discussed for both GPS and GLONASS. On one hand, some conventional error sources such as satellite orbit error, satellite clock error, ionospheric delay error, tropospheric delay error, receiver clock error, multipath error and noise need to be considered. On the other hand, some special error sources such as satellite and receiver antenna phase center offsets, relativistic effects, phase wind up, earth tide, ocean loading, atmosphere loading and Sagnac effect also need to be taken into account. Thirdly, traditional PPP model and P1-P2-CP PPP model are introduced in detail, and their corresponding stochastic models of observations and unknown parameters are discussed.The quality control method in the Kalman Filter processing is also given. Fourthly, since the existing PPP models could not process the combined GPS/GLONASS data due to the system time difference between GPS and GLONASS, the combined GPS/GLONASS PPP models have been developed, which include combined GPS/GLONASS traditional model and P1-P2-CP model.To assess the performance of the combined GPS/GLONASS PPP models, the observation data from IGS, precise satellite orbit and clock data from IAC (Information-Analytical Centre) are used to conduct numerical computation. The three-dimension positions, a receiver clock, a system time difference unknown and zenith wet tropospheric delay unknown are estimated and almost the same results are obtained from the combined GPS/GLONASS traditional model and P1-P2-CP model. Further investigation results obtained in static and kinematic modes indicate that adding GLONASS observations does not have a significant impact on positioning accuracy and convergence time when sufficient GPS satellites are available. In contrast, the positioning accuracy and convergence time could be significantly improved using the combined GPS/GLONASS PPP models when the number of GPS satellites is insufficient or GPS satellite geometry is poor.