Models and algorithms evaluation on GPS kinematic precise point positioning

In this paper, we will mainly concentrate on the models and the algorithms for kinematic precise point positioning (PPP). Based on a single station dual frequency data, a review of different combination of data types for PPP can be given, namely un-difference (UD), satellite difference (SD), time difference(TD) and time-satellite difference(TSD). Algorithms for solving the PPP observation models will also be presented. A discuss about the Kalman filtering, non-Kalman filter and some aspects about the convergence and reliability for kinematic PPP data process will be made. A three-day real data set will be used to evaluate the models and algorithms in this paper. A centimeter level precision for kinematic PPP from the float solution and a fixed solution after the ambiguity are hold as a fixed constant for post-process are given. According to the data process, some criterions such as the change rate of the estimated float ambiguity, the RMS of ambiguity and the change rate of the RMS for the ambiguity are adopted to evaluate the convergence of the float ambiguity in kinematic PPP. A preliminary criterion from the data process is given at the end of this paper.