北斗不同类型观测值随机模型精化  

Stochastic model refinement of different types of Beidou observations

作　　者：许振浩 高成发[1] 尚睿[1] XU Zhenhao;GAO Chengfa;SHANG Rui(School of Communications,Southeast University,Nanjing 211189,China)

机构地区：[1]东南大学交通学院,南京211189

出　　处：《测绘工程》2023年第6期38-43,53,共7页Engineering of Surveying and Mapping

摘　　要：针对BDS-2/BDS-3 GEO/IGSO/MEO卫星观测值质量存在差异的问题,对随机模型进行精化,以提高基线解算的精度。文中介绍随机模型的精化过程,统计不同类型卫星单差残差的时间序列,并分析其随高度角变化情况,基于最小二乘拟合法确定精化随机模型的参数。用不同长度的基线实验,比较精化模型与传统经验模型的定位精度。结果表明,精化模型能有效降低质量较差观测值的权重,提升定位精度和稳定性,相对于经验模型,E、N方向平面定位精度可提升10.9%~20%,高程定位精度可提升19.5%~31.1%。Aiming at the problem of differences in the quality of BDS-2/BDS-3 MEO/IGSO/GEO satellite observations,the stochastic model was refined to improve the accuracy of baseline solution.In this paper,the refining process of the stochastic model is introduced.The time series of single difference residuals of different types of satellites,and their changes with altitude angles are analyzed.The parameters of the refined random model are determined based on the least square fitting method.The positioning accuracy of the refined model is compared with that of the traditional empirical model by baseline experiments of different lengths.The results show that the refined model can effectively reduce the weight of poor quality observations,improve positioning accuracy and stability.Compared with the empirical model,the planar positioning accuracy can be improved by 10.9%~20%,and the elevation positioning accuracy can be improved by 19.5%~31.1%.

关 键 词：北斗 相对定位 随机模型 精度分析 

分 类 号：P228[天文地球—大地测量学与测量工程]