摘要
基于观测值域双差残差恒星日滤波方法消除多路径误差,利用事先计算的前一天观测卫星双差残差并进行去噪处理,根据各自卫星周日重复性改正到第二天对应时刻的双差观测方程中,得到剔除多路径误差后的"干净"双差观测值,最后解算基线信息。试验结果表明,与常规基于坐标域恒星日滤波相比,该方法在平面方向的精度可提高23%,且高程方向也有一定程度的提高。
Multipath must be eliminated as an important error source when improving relative positioning accuracy of GPS short baseline in practical deformation monitoring engineering such as for dams. high-rise buildings and big bridges. In this study, a new sideral algorithm based on a double residual in the GPS observation domain is proposed to mitigate mutipath errors. We calculated residuals for all double observations after resolving ambiguity while denoising those residuals for the first day. These results were taken as a multipath model to remove the mutipath errors from the double difference observation equations of the subsequent day by taking advantage of sidereal repeatability of every satellite. Clean double observations without multipath errors could be extracted, and a better baseline coordinate sequence was consequently calculated. The results show that this new method not only does well in the vertical precision but also reduces the root mean squre（rms） in horizontal coordinate precision by 23%, as compared with traditional sidereal filter .
出处
《武汉大学学报（信息科学版）》
EI
CSCD
北大核心
2014年第2期147-151,共5页
Geomatics and Information Science of Wuhan University
基金
国家自然科学基金资助项目(41074008)
高等学校博士学科点专项科研基金资助项目(20120141110025)
国家973计划资助项目(2012CB957701)
国家863计划资助项目(2012AA12A202)
武汉大学博士研究生自主科研项目基金资助项目(201161802020001)~~
