Orbits and Clocks for GLONASS Precise-Point-Positioning
Orbits and Clocks for GLONASS Precise-Point-Positioning
复制标题
DOI：
--
复制
发表时间：
2009-09
期刊：
影响因子：
--
通讯作者：
R. Piriz;D. Calle;A. Mozo;P. Navarro;D. Rodríguez;G. Tobías
中科院分区：
文献类型：
--
作者：
R. Piriz;D. Calle;A. Mozo;P. Navarro;D. Rodríguez;G. Tobías
2008年，GMV推出了MagicGNSS[参考文献1]，这是一个网络应用程序，提供了一套用于全球导航卫星系统数据处理的高精度和完整性工具。MagicGNSS的主要应用是GPS卫星轨道和时钟的计算，以及站/接收机坐标、对流层延迟和时钟的计算。MagicGNSS当前版本(1.3)可在http://magicgnss.gmv.com.上注册用户在线获取轨道确定和时间同步(ODTS)模块是MagicGNSS可用的第一个算法。ODTS通过处理分布在世界各地的GPS站网络的双频代码和相位测量来生成轨道和时钟。来自国际全球导航卫星系统服务处一组所谓核心站的过去和当前数据(国际导航卫星系统，[参考2])可在MagicGNSS服务器上使用，也可以从任何用户站上传和处理RINEX观测文件。自2009年初以来，俄罗斯GLONASS星座有19颗正在运行的卫星。因此，在MagicGNSS中除全球定位系统外还实施全球导航卫星系统数据处理，是利用扩大的卫星供应的一个自然步骤。GLONASS轨道和时钟的精确确定和预测带来了一些困难。其中之一是GLONASS卫星的先验未知太阳辐射模型，这实际上是亚分米轨道动力学精度的主要限制。此外，当处理来自GLONASS站网络的数据时，必须在站级别估计所谓的通道间偏差，以便通过站硬件和软件补偿GLONASS信号和代码的不同内部延迟。最后，俄罗斯卫星时钟的短期稳定性可能会给时钟估计和内插带来问题。精密单点定位(PPP)是一种相对较新的技术，可利用独立的全球导航卫星系统接收器和预先计算的精密卫星轨道和时钟产品(例如来自IGS的产品)，实现厘米级的定位误差和亚纳秒级的计时误差。用户接收器可以固定在地面(静态PPP)，也可以是漫游接收器(动态或动态PPP)。PPP与实时运动学(RTK)等其他精确定位方法的不同之处在于不需要基站或参考站。必须处理的唯一观测数据是用户接收器数据本身，因此降低了服务所需的带宽和计算功率。PPP的另一个优点是，由于输入的卫星轨道和时钟产品本质上是全球的，PPP解决方案也是全球的，即PPP方法适用于位于地球表面或上方的任何接收器，并且所产生的PPP解决方案(坐标)参考众所周知的地面参考框架(通常为ITRF)。MagicGNSS现已提供PPP软件模块。PPP模块处理GPS和GLONASS数据。本文从算法和可用性的角度描述了GLONASS数据处理在MagicGNSS网络应用程序的ODTS和PPP模块中的实现，并给出了在GLONASS轨道和时钟估计精度(ODTS)以及由此产生的用户接收器级定位和定时精度(PPP)方面的主要结果。对于PPP，考虑并报告了三种情况：仅GPS、仅GLONASS和GPS+GLONASS。分析了从1小时到1天的数据间隔。
In 2008 GMV introduced magicGNSS [Ref. 1], a web application providing a suite of tools for GNSS data processing featuring high-precision and integrity. The main application of magicGNSS is the calculation of GPS satellite orbits and clocks, and also of station/receiver coordinates, tropospheric delay and clock. magicGNSS current version (1.3) is available online for registered users at http://magicgnss.gmv.com. The Orbit Determination & Time Synchronization (ODTS) module was the first algorithm available in magicGNSS. ODTS generates orbits and clocks by processing dual-frequency code and phase measurements from a network of GPS stations distributed worldwide. Past and current data from a set of so-called core stations from the International GNSS Service (IGS, [Ref. 2]) is available on the magicGNSS server, and the possibility also exists to upload and process RINEX observation files from any user station. Since the beginning of 2009 the Russian constellation GLONASS has 19 operational satellites. The implementation of GLONASS data processing in addition to GPS in magicGNSS is then a natural step in order to take advantage of the extended satellite availability. The precise determination and prediction of GLONASS orbits and clocks poses a number of difficulties. One of them is the a priori unknown solar radiation model for the GLONASS satellites, actually the major limitation for sub-decimeter orbital dynamics accuracy. Also, when processing data from a network of GLONASS stations, a so-called inter-channel bias has to be estimated at station level in order to compensate for the different internal delays of the GLONASS signals and codes through the station hardware and software. Finally, the short-term stability of the Russian satellite clocks might pose a problem for clock estimation and interpolation. Precise-Point-Positioning (PPP) is a relatively new technique for centimeter-level error in positioning, and sub-nanosecond error in timing, using a stand-alone GNSS receiver and precise satellite orbit and clock products calculated beforehand (for example products from IGS). The user receiver can be fixed to the ground (static PPP) or be a roving receiver (dynamic or kinematic PPP). PPP is different from other precise-positioning approaches like Real Time Kinematics (RTK) in that no base stations or reference stations are needed. The only observation data that must be processed is the user receiver data itself, thus reducing the bandwidth and calculation power needed for the service. Another advantage of PPP is that since the input satellite orbit and clock products are by nature global, the PPP solutions are also global, i.e., the PPP approach works for a receiver located anywhere on or above the Earth surface, and the resulting PPP solution (coordinates) are referred to a well-known terrestrial reference frame (normally ITRF). A PPP software module is now available in magicGNSS. The PPP module processes GPS and GLONASS data. This paper describes the implementation of GLONASS data processing in the ODTS and PPP modules of the magicGNSS web application, from the point of view of both algorithms and usability, and presents the major results in terms of GLONASS orbit and clock estimation accuracy (ODTS), and in terms of the resulting positioning and timing accuracy at user receiver level (PPP). For PPP three scenarios are considered and reported: GPS-only, GLONASS-only, and GPS+GLONASS. Data intervals from 1 hour to 1 day are analyzed.