Algorithm Researching Of VRS And Implementing Based On The Multiple Reference Station Network

The network RTK, which is also called multi-base stations RTK, has more than three reference stations in one area and covers one area with network, provides real time error correction for users through make one or more stations as datum to calculate and broadcast correction information. It overcome the weakness of traditional RTK technique, enlarged the distance of RTK surveying, and improved positioning result and reliability. VRS technique is a kind of positioning method integrating the internet, wireless communication technique, computer network, data base and GPS positioning, and represents a novel developing trend of GPS.The emphasis is positioning algorithm for VRS technique. Firstly, the thesis introduces the traditional RTK system and the network RTK system, analyses the merits and applications of network RTK through of comparing with them, and introduces the network system in details which includes principles, classifications, component parts and data flow. The GPS observations in common use, equation for observation and their linear combination are also introduced. Secondly, the thesis analyzes all kinds of errors and eliminating method, discusses ionosphere error and troposphere error on emphasis, and educes the formulas for positioning using of VRS which includes three steps: the generation of correction between reference stations, the generation of virtual observations and the positioning calculation in rover station. Aiming at characteristics of network RTK, a simple method for fixing ambiguity is provided between reference stations. Finally, the thesis researches on several interpolated methods for VRS in common use, analyzes their merits and shortcomings and provided proving methods.For eeperiment, the data gets from the general verification network for GPS receivers and a VRS network. Computed through of compiling programs, the positioning results achieve centimeter class. The mean square error on X-coordinate is 1.6cm. The mean square error on Y-coordinate is 2.1cm. The mean square error on Z-coordinate is 1.5 cm. So the VRS algorithm is attested to be correct and feasible in the paper.