基于电离层延迟智能估计的完好性评估  

Integrity evaluation based on intelligent estimation of ionospheric delay

作　　者：卢丹 张宏健 钟伦珑 胡铁乔 LU Dan;ZHANG Hongjian;ZHONG Lunong;HU Tieqiao(College of Electronic Information and Automation,CAUC,Tianjin 300300,China)

机构地区：[1]中国民航大学电子信息与自动化学院,天津300300

出　　处：《中国民航大学学报》2025年第1期53-59,共7页Journal of Civil Aviation University of China

基　　金：国家重点研发计划项目(2020YFB0505603)。

摘　　要：完好性是民用航空导航系统的性能要求之一,而电离层延迟是进行完好性评估的重要误差来源之一。对于单频系统来说,通常采用Klobuchar模型对单频信号进行修正,但修正率仅为60%左右,精度有限。本文提出一种估算电离层延迟并将其用于完好性评估的方法,首先,通过反向传播(BP,back propagation)神经网络对观测站电离层延迟进行训练,得到观测站的BP神经网络模型并计算电离层延迟,提高了观测站的电离层延迟修正率;其次,利用经过训练的观测站BP神经网络模型以及三点插值法计算装备单频接收机的航空器在其飞行位置的电离层延迟并进行完好性评估。以2015年2-3月的数据为例进行评估,结果表明,通过BP神经网络模型以及三点插值法计算部分地区的电离层延迟与Klobuchar模型相比,计算电离层延迟效果较好,提高了日间和夜间电离层延迟的修正率,从而提高了保护级的精度。Integrity is one of the performance requirements of civil aviation navigation systems,and ionospheric delay is an important error source for integrity evaluation.For the single-frequency system,Klobuchar model is usually used to correct the single-frequency signal,but the correction rate is only about 60%,and the accuracy is limited.This paper proposes a method for estimating ionospheric delay and using it for integrity evaluation.Firstly,the iono-spheric delay of observation station is trained by back propagation(BP)neural network,and the ionospheric delay is calculated by BP neural network model of the observation station,which improves the ionospheric delay correc-tion rate of the observation station.Secondly,the ionospheric delay of the aircraft equipped with single-frequency receiver at its flight position is calculated by using the trained observation station BP neural network model and the three-point interpolation method,and the integrity evaluation is carried out.Evaluation based on the data from February to March 2015 shows that,when calculating the ionospheric delay,the BP neural network model and the three-point interpolation method perform better than the Klobuchar model in some areas.They improve the cor-rection rates of ionospheric delay during both daytime and nighttime,thus enhancing the accuracy of the protec-tion level.

关 键 词：电离层延迟 完好性评估 KLOBUCHAR模型 BP神经网络 保护级 

分 类 号：P228[天文地球—大地测量学与测量工程]