The System Analysis And Implementation Of GPS Long Baseline Data Processing Software

GPS long baseline data processing software is a very complex system. As high-precision data processing software, CGO simplifies a lot of parameter settings and operating steps, reducing the difficulties when processing GPS data. However, several subtle issues are yet remained in the previous version. This study aims at integrating and optimizing part of the system, such as ambiguity fixed rate, baseline accuracy, adjustment module, processing speed and auxiliary function modules, sourcing from the theory of high-precision GPS data processing. In order to solve these problems, this article proposes a set of solutions based on the old platform. The detail information is summarized as follows:1:The Three-dimensional unconstrained adjustment, Three-dimensional constraint adjustment and Two-dimensional constraint adjustment are developed based on the old platform. The combination of network adjustment module and baseline resolution module in the new system makes a practical application of GPS data processing.2:More feasible auxiliary modules rich the users’experience of using CGO, such as coordinate transformation, time conversion, Google Maps, data preprocessing and so on.3:Several experiments (short baseline, middle-long baseline, long baseline and ribbon hybrid network) comparing with mainstream GPS software (GAMIT, COSA_GPS) are designed to verify CGO’s reliability.4:A series of experiment results have proved the way how the determined factors would affect the positioning precision. First of all, we conclude that rapid ephemeris can replace precise ephemeris in the case of non-specific condition. Secondly, we find that the antenna’s PCV can affect the final positioning in different ways. It has a greater impact on the long baseline than short baseline. However, there is nearly no effect on the short baseline when using the same antenna. More than that, when researching the elevation cutoff’s effect on baseline resolution, we discover10degrees is better than5and25. Fourthly, by discussing the influence of time on the final position, we find that accuracy of positioning tends to be stable when observation time up to8hours. Fifth, when studying the effect of observation session on baseline resolution, we find accuracy of baseline is slightly better when observation session is in the evening. Sixth, by discussing the sampling rate’s effect on the final position, we find that the results are almost unanimously for30and60second sampling rate.