静基座全球定位系统定位误差分析和建模研究

Analyzing and Modeling of GPS Positioning Data on Static Bases

本文以静基座全球定位系统（ＧＰＳ）实测定位数据为依据，分析其定位误差情况，并进行建模研究．研究表明，静基座ＧＰＳ陆定位误差模型国定位星组的不同而有差异，但都含有周期为１８０～６００５幅值为随机变量的周期误差和符合ＡＲ（１）模型的随机误差．木文采用半周期均值法和ＡＲ模型预报法对定位数据进行误差补偿．

Based on the actual GPS data on static bases, the error of positioning is analyzed in this paper. It is shown that the accuracy of GPS positioning is 100m (two-dimentional radical error, 95th pencentile). The error of positioning jumps when the set of positioning satellites is changed. The error models which are set with these tiPS data are different with the set of positioning satellites. But they all consist of a frequency error, with a frequency of 3 to 10 minutes within a random range, and an AR(1) random error. The models are used with the half frequency mean method and the AR model forecast method to compensate forthe positioning error so that the accuracy of the SPS by GPS is improved.