Modeling And Precision Analysis Of Atmosphere Based On GNSS

Ionosphere and troposphere, linked inextricably with our living and production nowadays, on which many explorations and studies were made in the early nineteenth century, are significant components of geospace.In recent years, GNSS(Global Navigation Satellite System) was widespreadly used and the researches and applications of ionosphere-related or troposphere-related science and technology were bound together with the influence of radio(GNSS signals) propagation in geospace. For this reason, the paper, based on actual GNSS observation, accomplished such following works focusing on ionosphere and troposphere.In the ionosphere respect,this paper presents a constrained adaptive simultaneous iteration reconstruction technique to improve the deficiencies of traditional CIT(Computerized Ionospheric Tomography)algorithm.And numerical test and actual GNSS observation experiment validates the feasibility and superiority of the new method, In the troposphere respect, this paper, utilizing mass actual GNSS observation supplied by IGS(International GNSS Service), analyze the precision and adaptability of ZTD(Zenith Tropospheric Delay) model and on account of its shortcoming presents a new ZTD predicted model, in additions, analyze and verify the new model's feasibility according to test result. Detailed works and results are such following as:1)Adaptive simultaneous iteration reconstruction technique is a fast algorithm for the tomographic reconstruction of ionospheric electron density distribution.Howerver,the algorithm is sensitive to the initial value,which are obtained from empirical ionospheric model.To resolve the above problem,a constrained adaptive simultaneous iteration reconstruction technique is presented in this paper.The new method imposes reasonable constrains on the tomography system by considering the continuousness of neighboring cells,and the accuracy of ionospheric electron density reconstruction is efficiently improved. Numerical test and actual GNSS observation experiment validates the feasibility and superiority of the new method.2)This paper makes use of ZTD data in the year 2014 of the 36 stations(IGS) to analyze the accuracy and adaptability of the existing typical models including Hopfield,Saastamoinen taked as the representative containing meteorological parameters, and SHAO model as no meteorological parameters.The results indicate if containing meteorological parameters,the precisions of the four models are able to reach cm level,if not,the others comparing with SHAO are degraded and are poorly appropriate to the correction of tropospheric delay,especially,Hopfield is not sufficient accuracy in high altitude to estimate ZTD of that zone.3)Considering the shortcomings of above existing models,this paper, based on actual GNSS observation,presents a new method to model and predict tropospheric delay with no need for meteorological parameters. Firstly, the cycle characteristics of time series of ZTD are obtained using the spectral analysis. On the above basis, the ZTD model is created and the calculation method of model parameters is given. Secondly, the value of ZTD is fitted based on spectral analysis. A comparison is made between the fitting values and the actual data. And the fitting residual is gotten. Finally, the fitting residual is corrected and predicted using the Auto Regressive model(AR model). The results show that the corrected and predicted model of ZTD based on spectrum analysis and the AR compensation can meet the needs of different testing environment, and the accuracy could reach centimeter level.