GPS/BDS卫星姿态异常对PPP相位缠绕的影响及其改正模型

The Impact of Yaw Attitude of Eclipsing GPS/BDS Satellites on Phase Wind-up Solutions for PPP and Its Correction Model

在精密单点定位中,相位缠绕是一项不可忽略的误差。相位缠绕的计算严格依赖于卫星姿态的确立,不同的卫星类型产生不同的异常。本文给出了卫星在正常情况下的姿态模型和在异常情况下的姿态改正模型。使用真实数据测试以验证本文所提出模型的正确性。观察滤波收敛后出现异常情况的卫星观测值的残差,结果表明:在异常时期残差最大可能超过20 cm,然而使用本文的改正模型,残差可降低到5 cm以下。使用不同分析中心的精密轨道和钟差产品,效果存在微小差异。II/IIA卫星通过地影区域的时间最长可达1 h,此期间卫星姿态完全受航向角偏差(II/IIA为+0.5°)控制,出了地影区域后30 min,姿态难以模型化,因此这30 min的观测数据不建议采用。

Care of the phase wind-up correction should be reasonably taken in precise point positioning.In practice,correct computation of phase wind-up relies mainly upon the information about the satellite attitude,which should be modeled differently when satellites undergo eclipsing.Different GPS satellite types would be subject to different eclipsing periods.For instance,GPS IIR satellites can experience noon and midnight turn maneuvers,GPS IIF satellites suffer from noon maneuver and shadow crossing,and GPS II/IIA satellites may further experience post-shadow recovery periods when compared to IIF ones.As for the BDS non-GEO satellites,one should take into account the attitude control switching between the nominal and the orbit-normal mode.This paper presents a model enabling the attitude to be correctly computed for both eclipsing as well as non-eclipsing satellites.Numerical tests using real data are then performed in order to verify our model presented.As far as the filtered residuals are concerned,it is found that,their maximum residual could exceed 20 cm during the eclipsing periods.This problem is fortunately solvable when use of our model has been made,since the residuals reduce to below5 cm.It should be noted that,our numerical results may be slightly different when we use precise satellite orbit and clock products delivered by different Analysis Centers.Furthermore,the shadow crossing period takes typically up to 1hour for GPS II/IIA satellites,during which the yaw attitude is controlled entirely by the positive yaw bias（II/IIA of＋0.5°）.The II/IIA post-shadow recovery periods,covering about 30 minutes,still cannot be fully modeled;the data collected within this period should thereby be excluded.