测绘学报

• 学术论文 • 上一篇    下一篇

地球静止轨道卫星广播星历参数拟合与试验分析

何峰   

  1. 北京环球信息应用开发中心
  • 收稿日期:2011-03-28 修回日期:1900-01-01 出版日期:2011-05-05 发布日期:2015-06-24
  • 通讯作者: 何峰

Ephemeris Fitting and Experiments Analysis of GEO Satellite

  • Received:2011-03-28 Revised:1900-01-01 Online:2011-05-05 Published:2015-06-24

摘要: 我国的区域卫星导航系统COMPASS中,广播星历利用考虑一阶摄动影响的15个轨道根数参数描述卫星轨道,系统导航电文中给每个参数分配的比特位与GPS相同。然而,由于GPS星座由周期约12小时的MEO卫星构成,其轨道参数化方案和每个参数的比特位分配方案可能不适用于更高轨道的GEO和IGSO卫星。COMPASS的系统设计中考虑了由于GEO的小倾角、小偏心率特点产生的广播星历拟合精度损失问题,然而在系统组网运行过程中进一步发现在克服精度损失问题后,在特定时段,GEO广播星历部分参数在编码导航电文时超过了系统设计分配给该参数的比特位限制,即参数超限。详细介绍GEO卫星广播星历参数超限的现象和规律,根据解算方程病态性和轨道动力学特性分析了该参数超限的原因,设计了在不显著损失精度前提下多种改善参数超限的拟合方案,利用实测数据进行了试验。结果表明:固定超限参数的同时解算其余星历参数的方法和进行参数岭估计的方法,都能够有效地抑制并解决参数超限问题。

Abstract: The satellite orbit is described by 15 parameters considering the one scale perturbation in COMPASS, and the bit distribution of navigation message is also consistent with GPS. But because the constellation of GPS is consisted of MEO whose period is approximately 12 hours, the parameterized program and bit distribution program may not applicable to GEO and IGSO which have higher orbit. In the system design of COMPASS, the ephemeris parameters accuracy losing have been taken into consideration, which is caused by the little inclination and small eccentricity of GEO. In the operation of system, thought the accuracy losing is eliminated, in special intervals, some parameters exceed the distributed bit limitation when the ephemeris parameters are coded into navigation message. The phenomenon and regulation of parameter’s exceeding limitation in the operation of system are introduced, and then the reasons are illustrated according to the ill-conditioned equation and the orbital characteristics, Some fitting algorithms are designed on the condition of avoiding accuracy losing, and the experiments are carried out with the actual observation data. The results proves, it is efficient to eliminate the parameter exceeding for the method of fixing one parameter while computing the others or the method of parameter ridge estimate.