Abstract

As part of the International GNSS Service (IGS), several analysis centers provide GPS and Galileo satellite phase bias products to support precise point positioning with ambiguity resolution (PPP-AR). Due to the high correlation with satellite orbits and clock offsets, it is difficult to assess directly the precision of satellite phase bias products. Once outliers exist in satellite phase biases, PPP-AR results are no longer reliable and the combination of satellite phase bias products from IGS analysis centers also gets difficult. In this contribution, we propose a method independent of ground measurements to detect outliers in satellite phase biases by computing the total Difference of satellite Orbits, Clock offsets and narrow-lane Biases at the midnight epoch between two consecutive days. Results over 180 days show that about 0.2, 1.1, 2.0 and 0.1% of the total DOCB values for GPS satellites exceed 0.15 narrow-lane cycles for CODE final, CODE rapid, CNES/CLS final and WUHN rapid satellite products, respectively, while the same outlier-ratios for Galileo satellites are 0.1, 0.9, 0.4 and 0.1%, respectively. As an important contribution to the orbit, clock and bias combination task, we check the consistency of satellite phase bias products between two analysis centers before and after removing these detected outliers from individual analysis centers. It is convincing that the number of large differences of satellite phase biases between two analysis centers is notably reduced.

中文翻译：

---

一种评估GNSS卫星相位偏差产品质量的方法

摘要

作为国际 GNSS 服务 (IGS) 的一部分，多个分析中心提供 GPS 和伽利略卫星相位偏差产品，以支持具有模糊度分辨率的精确单点定位 (PPP-AR)。由于与卫星轨道和时钟偏移的高度相关性，直接评估卫星相位偏差产品的精度很困难。一旦卫星相位偏差存在异常值，PPP-AR结果就不再可靠，并且IGS分析中心的卫星相位偏差产品的组合也变得困难。在本文中，我们提出了一种独立于地面测量的方法，通过计算连续两天之间午夜时段的卫星轨道、时钟偏移和窄道偏差的总差来检测卫星相位偏差的异常值。 180 天的结果显示，对于 CODE Final、CODE Rapid、CNES/CLS Final 和 WUHN Rapid 卫星产品，GPS 卫星总 DOCB 值的约 0.2%、1.1%、2.0% 和 0.1% 超过 0.15 个窄带周期，而伽利略卫星的相同异常值比率分别为 0.1%、0.9%、0.4% 和 0.1%。作为对轨道、时钟和偏差组合任务的重要贡献，我们在从各个分析中心去除这些检测到的异常值之前和之后检查两个分析中心之间卫星相位偏差产品的一致性。令人信服的是，两个分析中心之间卫星相位偏差差异较大的数量显着减少。