Regional Ionospheric Modeling Of LEO Constellation-augmented GNSS

As a part of the near-earth space environment,the ionosphere has an important influence on the production and life of human beings.Ionospheric delay error is a very important error source in GNSS positioning system,which seriously affects the positioning accuracy of GNSS,especially that of the single frequency positioning users.At present,the LEO satellites(Low Earth Orbit,LEO)have entered into a period of vigorous development and the LEO satellite constellation era is coming.In addition to realizing the navigation enhanced positioning function,the LEO satellites planned for launching can also receive LEO navigation enhanced signals and GNSS satellite signals by taking advantage of ground stations and LEO satellite-borne receivers respectively to achieve the ionosphere detection of the lower and upper parts of LEO satellites.This provides a new opportunity for the establishment of high-precision ionospheric models.The new ionospheric observations of LEO satellites can not only provide extremely rich observational data for ionospheric modeling,but also bring new challenges at the same time.This paper aims to study regional ionospheric modeling and monitoring of LEO constellation-augmented GNSS.Based on the simulated GNSS and LEO satellite constellation,the ionospheric observation data of GNSS/LEO is simulated,the contribution of LEO satellites to ionosphere observation is analyzed and the ionospheric modeling accuracy that both considers and disregards all kinds of errors and the accuracy of Differential Code Bias(DCB),a byproduct of modeling,are evaluated.The main contents and conclusions of this paper are listed below:(1)The current status of construction and development of low orbit satellites,the research status of regional ionospheric modeling at home and abroad and the basic methods and modeling principles of ionospheric detection at present are systematically introduced.The projection function,empirical model and mathematical model in the process of ionospheric modeling are summarized.(2)The ionospheric observation data in the African region that both considers and disregards all kinds of errors are simulated.The simulated ionospheric observation data contain three types: the observation data of GNSS satellites from ground stations(abbreviated to Station2GNSS),the observation data of LEO satellites from ground stations(Station2LEO)and the observation data of GNSS satellites from LEO satelliteborne receivers(abbreviated to LEO2GNSS).Since the data of the latter two are the observation data of the lower and upper parts of LEO respectively,according to the scaling factor derived from the calculation based on IRI(International Reference Ionosphere)model,they are converted into the observation data of the complete path.This process is briefly called "normalization".(3)The distribution of Ionospheric Pierce Points(IPPs)of different LEO constellations and GNSS satellites is deeply analyzed.The results show that: when LEO and GNSS satellites are observed from ground stations,the same LEO satellite can be observed twice within two hours,while the same GNSS satellite can only be observed once,and elevation angles and azimuth angles of LEO satellites vary more significantly over time than those of GNSS satellites;the IPPs of LEO satellites are better distributed in the African region without ground stations with north latitude 15°and east longitude20° as the center;the number of orbital planes of LEO constellation affects the distribution of IPPs,that is,the distribution of IPPs of Station2 LEO and LEO2 GNSS with 6 orbits is significantly different from that with 12 orbits.(4)The influence of LEO constellation on regional ionospheric modeling while disregarding all kinds of errors is analyzed,and the model accuracy analysis for different modeling durations,constellation types and latitudes is completed.The results show that: as for the fusion ionospheric modeling results of GNSS and different LEO constellations,with the increase of the number of LEO satellites,the accuracy of ionospheric modeling is constantly improved;with the increase of observation time,the modeling difference between GNSS+LEO and single GNSS decreases gradually;at the location of east longitude 30° and north latitude 10°,the maximum differences of the modeling results between GNSS+288LEO and single GNSS and between GNSS+192LEO and single GNSS are-1.6 TECU and-1.4 TECU respectively;in the region of central Africa without measurement stations,the difference between the GNSS+LEO and single GNSS modeling results in 15 minutes is relatively large,indicating that the observation with the addition of LEO satellites has a significant contribution to ionospheric modeling.(5)The influence of the addition of LEO satellites on regional ionospheric modeling when taking into account of various errors is studied.The accuracy assessment of difference code deviation at the satellite end and receiver end of each system and the accuracy analysis of fusion modeling are completed.For the accuracy assessment of DCB at the satellite end,the magnitude of Bias and RMS of DCB of LEO satellites is larger than that of GNSS satellites.The estimation accuracy of DCB of GNSS satellites is closely related to IPPs distribution of GNSS satellites observed from the ground,and the estimation accuracy of DCB of satellites with poor IPP distribution is obviously worse.For the accuracy assessment of DCB at the measuring station end,the Bias and RMS of DCB at the receiver ends of GNSS and LEO are both within 2ns and the Bias and RMS of DCB of LEO satellite-borne receiver are both within 0.5 ns.Experimental results show that: the fusion modeling scheme is superior to the single GNSS observation data modeling scheme,and the RMS value of the former is significantly lower than that of the latter;compared with the single GNSS modeling scheme,the STD and RMS of fusion modeling improve by 23.75% and 23.71%respectively to the largest extent,and the effect is very significant.(6)The effects of different "normalized" cut-off elevation angles and different LEO2 GNSS median errors on regional ionospheric fusion modeling are investigated.The results show that the accuracy of fusion modeling decreases significantly with the increase of "normalized" cut-off height Angle,but it is still better than single GNSS modeling.The accuracy of fusion modeling increases slightly with the decrease of the median error.Among the 12 fusion modeling schemes,the fusion modeling accuracy is highest when the median error of LEO2 GNSS is 0.25? and the "normalized" cut-off elevation angle is 20°.