曹士龙, 刘根友, 王生亮, 高铭, 尹翔飞. GPS超长基线解算的误差特性与精度分析[J]. 武汉大学学报 ( 信息科学版), 2023, 48(2): 260-267. DOI: 10.13203/j.whugis20200322
引用本文: 曹士龙, 刘根友, 王生亮, 高铭, 尹翔飞. GPS超长基线解算的误差特性与精度分析[J]. 武汉大学学报 ( 信息科学版), 2023, 48(2): 260-267. DOI: 10.13203/j.whugis20200322
CAO Shilong, LIU Genyou, WANG Shengliang, GAO Ming, YIN Xiangfei. Bias Characteristics and Accuracy Analysis of GPS Ultra-Long Baseline Solution[J]. Geomatics and Information Science of Wuhan University, 2023, 48(2): 260-267. DOI: 10.13203/j.whugis20200322
Citation: CAO Shilong, LIU Genyou, WANG Shengliang, GAO Ming, YIN Xiangfei. Bias Characteristics and Accuracy Analysis of GPS Ultra-Long Baseline Solution[J]. Geomatics and Information Science of Wuhan University, 2023, 48(2): 260-267. DOI: 10.13203/j.whugis20200322

GPS超长基线解算的误差特性与精度分析

Bias Characteristics and Accuracy Analysis of GPS Ultra-Long Baseline Solution

  • 摘要: 基于单基站的超长基线定位技术在地壳形变监测、高精度授时等领域具有广泛应用,但仍有诸多因素制约着超长基线解算精度。从观测方程出发,利用单差观测值对长(超长)基线(146~1 724 km)解算中的卫星轨道误差、对流层延迟误差、地球潮汐误差和相位缠绕误差等误差特性进行了详细分析。分析结果表明,当基线小于500 km时广播星历误差可忽略不计;超过500 km时需要采用精密星历, 同时需要考虑地球潮汐误差的影响;利用参数估计法同时估计基线两端的天顶对流层延迟误差可获得1~2 cm精度; 相位缠绕误差对基线小于2 000 km的解算影响可忽略。基于估计天顶对流层延迟的方法解算了5条长(超长)基线(146 km、491 km、837 km、1 043 km和1 724 km)。实验结果表明,当基线小于500 km时,采用广播星历可获得水平方向优于0.05 m、高程方向优于0.08 m的定位精度;当基线小于2 000 km时,采用超快速精密星历可获得水平方向优于0.025 m、高程方向优于0.055 m的定位精度。解算的初始收敛时间随着基线长度增加而缩短。

     

    Abstract:
      Objectives  Considering the advantages of low cost, simple algorithm model and high accuracy, the ultra-long baseline positioning technology was widely used in the fields of crustal deformation monitoring and high-precision timing. However, there are still many factors that may reduce the accuracy of the ultra-long baseline solution, such as satellite orbit error, tropospheric delay error and solid earth tide, etc.
      Methods  In this contribution, characteristics of various errors and bias (such as broadcast ephemeris orbit error, tropospheric delay error, earth tide and phase windup) were analyzed based on ionosphere-free combination model. The rule of error changing with baseline length was analyzed using single difference observations of long(ultra-long) baselines (146-1 724 km).
      Results  The results show that broadcast ephemeris error could be neglected while baseline is less than 500 km. When baseline length exceeded 500 km, precise ephemeris was needed to reduce the influence of orbital errors on positioning. The accuracy of the zenith tropospheric delay calculated by GPT2+Saastamoninen model ranged from centimeter to decimetre, and the method of simultaneously estimating zenith tropospheric delays at both stations could provide accuracy of 1-2 cm. Compared with the horizontal direction, the vertical direction of long baseline was more significantly affected by solid Earth tides which must be corrected in the solution of long baseline. Phase wind-up error could be neglected while baseline is less than 2 000 km.
      Conclusions  The experimental results of five long(ultra-long) baselines (146 km, 491 km, 837 km, 1 043 km and 1 724 km) solutions were presented using estimating zenith tropospheric delays at both stations of baseline. While baseline is less than 500 km, the positioning precision based on broadcast ephemeris is better than 0.05 m and 0.08 m in horizontal and vertical direction, respectively. While baseline is less than 2 000 km, the positioning precision based on ultra-rapid precise ephemeris is better than 0.025 m and 0.055 m in horizontal and vertical directions, respectively. The initial convergence time of the baseline solution decreased as the lengths of baseline increased.

     

/

返回文章
返回