基于神经网络的北京地区电离层TEC经验模型研究

Research on Ionospheric TEC Empirical Model in Beijing Based on Neural Network

利用2009—2019年间的全球电离层图(Global Ionospheric Map,GIM)产品,提出了一种基于神经网络技术的北京地区电离层总电子含量(Total Electron Content,TEC)经验模型建立方法。模型精度验证结果表明,本文提出的方法能够有效提高电离层TEC经验模型精度,相对Klobuchar模型和BDGIM模型精度分别提高了62%和21%。BJFS站的单频定位验证结果表明,本文建立的电离层TEC经验模型能够帮助单频用户有效提高定位精度,相对Klobuchar模型和BDGIM模型的三维定位精度分别提高了32%和7.5%。

This paper proposes a establishing method of empirical model for total electron content(TEC) in the Beijing region based on neural network technology using the Global Ionospheric Map(GIM) products from 2009 to 2019.The model accuracy verification results show that the proposed method can effectively improve the accuracy of the ionospheric TEC empirical model,which is 62% and 21% higher than the Klobuchar model and BDGIM model,respectively.The accuracy verification results of single frequency positioning using BJFS station show that the ionospheric TEC empirical model established in this paper can effectively improve the positioning accuracy of single frequency users by 32% and 7.5% compared to the 3D positioning accuracy of Klobuchar model and BDGIM model,respectively.