测绘通报 ›› 2018, Vol. 0 ›› Issue (5): 20-24.doi: 10.13474/j.cnki.11-2246.2018.0137

• 学术研究 • 上一篇    下一篇

EEMD分解的超快速北斗卫星钟差预报

毛亚1, 王潜心1, 胡超1, 张铭彬1, 于伟宣2   

  1. 1. 中国矿业大学环境与测绘学院, 江苏 徐州 221116;
    2. 武汉大学遥感信息工程学院, 湖北 武汉 430079
  • 收稿日期:2017-09-13 修回日期:2017-11-30 出版日期:2018-05-25 发布日期:2018-05-31
  • 作者简介:毛亚(1994-),男,硕士生,主要从事卫星钟差预报等的研究。E-mail:maoya0428@foxmail.com
  • 基金资助:

    国家自然科学基金(41404033);国家重点实验室开放基金重点项目(SKLGIE2014-Z-1-1)

Prediction of Ultra-rapid Ephemeris Clock Error Based on EEMD

MAO Ya1, WANG Qianxin1, HU Chao1, ZHANG Mingbin1, YU Weixuan2   

  1. 1. School of Environment Science & Spatial Informatics, China University of Mining and Technology, Xuzhou 221116, China;
    2. School of Remote Sensing and Information Engineering, Wuhan University, Wuhan 430079, China
  • Received:2017-09-13 Revised:2017-11-30 Online:2018-05-25 Published:2018-05-31

摘要:

为提高北斗卫星钟差预报精度,本文提出采用多项式和集合经验模态分解相结合的模型进行北斗卫星钟差的预报,并采用GPS数据验证算法的正确性。在剔除卫星钟差中的趋势项部分后,利用经验模态分解法对残差分解得到不同频率的时间序列。对各时间序列用不同模型进行预报并进行线性组合,最终的钟差预报值由趋势项和各时间序列的预报值复合而成。试验表明:该模型对北斗卫星钟差预报取得了较好的结果,与ISU-P相比,精度提升幅度在7.3%~43.0%之间。

关键词: 钟差预报, 集合经验模态分解, ISU-P, 抗差估计

Abstract:

In order to improve the prediction accuracy of BeiDou satellite clock offset, a module is applied to predict the clock offsets of BeiDou satellites using polynomial and EEMD. And the correctness of the module is verified by GPS data. After removal of trend items, obtaining the time series with different frequencies by decomposed residuals using EEMD. On the linear combination of the prediction value by different module, and the final clock offsets prediction value is combined by prediction value of the trend and the sub-time series. The experiment shows that, the module has achieved a good results in predicting BeiDou satellites clock offsets, and the accuracy increases between 7.3%~43.0% compared with ISU-P.

Key words: clock error prediction, EEMD, ISU-P, robust estimate

中图分类号: