【摘要】 针对GNSS高精度定位观测数据的可信程度量化评估问题，该文基于优劣解距离法(TOPSIS),提出了一种综合赋权的GNSS观测数据可信度评估方法。基于多路径效应、数据完整率、周跳以及信噪比等指标设计评估函数，进行指标的同趋势化以及标准化等数据预处理步骤后，利用熵权法计算各指标的客观权重，结合专家经验给出的主观标准权得到各测站观测数据好的综合可信评估结果。实验选用了我国四川省8个CORS站及其周边5个监测站进行可信度评估，比较了不同数据预处理方案对于评估结果的影响，结果表明采用区间化方法进行指标的标准化更符合实际。从稳定性和可用性角度分析比较了熵权TOPSIS法、标准权TOPSIS法和本文所述综合权TOPSIS方法，本文提出的综合权TOPSIS方法评估可信度稳定性最高，可用性更强。更多还原

【Abstract】 Regarding the quantitative evaluation of the credibility of GNSS high-precision positioning observation data, this paper proposes a comprehensive weighting method based on TOPSIS for evaluating the reliability of GNSS observation data. The evaluation function is designed based on indicators such as multipath effects, data integrity, cycle slips, and signal-to-noise ratio. After performing data preprocessing steps such as trend alignment and standardization of these indicators, the entropy weight method is used to calculate the objective weight of each indicator. Combining these with the subjective weights provided by expert experience, a comprehensive reliability evaluation result for observation data at each station is obtained. This paper selected 8 CORS stations and 5 surrounding monitoring stations in Sichuan Province, China, for the reliability evaluation, comparing the impact of different data preprocessing schemes on the evaluation results. Experimental results show that the standardization method using intervals aligns better with reality. Subsequently, the stability and usability of the entropy weight TOPSIS method, the standard weight TOPSIS method, and the proposed comprehensive weight TOPSIS method are analyzed. Experiments indicate that the proposed comprehensive weight TOPSIS method provides the highest stability and usability in evaluating reliability.更多还原