一种基于组合模型的接收机钟差预报  

A receiver clock bias forecast based on combined model

作　　者：孙凯[1,2] 蔡成林[1,2] 杨国刚[1,2] 吴国增[1,2] SUN Kai CAI Cheng-lin YANG Guo-gang WU Guo-zeng(Guilin University of Electronic Technology, Guilin 541000, China Guangxi Key Laboratory of Precision Navigation Technology and Application, Guilin 541000, China)

机构地区：[1]桂林电子科技大学,桂林541000 [2]广西精密导航技术与应用重点实验室,桂林541000

出　　处：《时间频率学报》2017年第2期97-104,共8页Journal of Time and Frequency

基　　金：国家自然科学基金资助项目(61263028)

摘　　要：在城市或者峡谷等地,可视卫星的数量急剧减少,卫星导航信号遮挡严重,研究小于4颗卫星定位方法具有重要意义。当只能观测到3颗或者2颗卫星时,可通过建立接收机钟差的组合预报模型对钟差进行短期预报从而降低定位方程维数,达到定位环境恶劣的条件下定位结果的连续性和可用性的目的。该组合模型首先对接收机钟差进行二项式拟合,然后对拟合之后的残差进行谱分析,最后通过ARIMA模型预报接收机钟差。通过实测数据验证得出:该组合预报模型能实现对接收机钟差的高精度短期预报且无需增加外设,对GNSS接收机来说具有很高的实用价值。The number of visible satellites can be significantly decreased under the circumstance while the receivers are in the urban areas or valley where the satellites signal may be blocked seriously. Thus, the scenario that the receivers can only observe three or two satellites to locate their positions is considerable valued in the practical application. In this paper, a proposed scheme that establish a combined forecasting model for short-term forecasts of receiver clock bias can reduce the dependence on the number of dimensions of equation, to achieve the purpose of continuity and availability of positioning results in the extremely conditions. First of all, the receiver clock bias was fitted by binomial, and then the fitting residual was processed by spectrum analysis, finally the receiver clock bias was predicted by ARIMA model. Experiments based on the measured data have demonstrated that this method can achieve high precision short-term forecasts for receiver clock bias without additional peripherals, which can be highly practical value for GNSS receivers.

关 键 词：接收机钟差 组合预报模型 短期预报 

分 类 号：P228[天文地球—大地测量学与测量工程]