【作者】 林乐科；

【导师】 张业荣； 赵振维；

【作者基本信息】 南京邮电大学 ， 电磁场与微波技术， 2011， 博士

【摘要】 对流层的折射率高度剖面是电波折射修正的基础数据,但是目前还缺乏较为简便的全天候实时获取方法。全球导航卫星系统(GNSS)不仅成功应用于测距、授时和导航等传统领域,而且在环境监测、大地测量、天文、气象等领域也得到广泛的应用。本文基于单站地基GNSS估计的电波折射参数(延迟与弯曲角,以延迟为主),研究了大气折射率剖面的反演算法。主要研究成果如下:1)传统的GNSS高精度单点定位(PPP)方法获取对流层延迟是基于IGS精密星历实现的,但是精密星历的发布有13天的延迟,因此对流层延迟不能够实时获取。本文在GNSS高精度单点定位(PPP)方法和IGS超快速星历基础之上,提出了单站地基GNSS接收机对低高度角(<5°)对流层斜延迟的实时估算方法,实验证实该方法获取的斜延迟误差小于5%。2)基于大气折射率剖面模型和GNSS探测的对流层延迟,提出了基于GNSS的对流层折射率模型参数的估算反演方法。这种反演方法的主要优点是不需要历史数据。3)通过对历史探空数据及其GNSS信号延迟仿真的分析,表明不同高度的折射率与地面气象参数和对流层延迟之间有很好的相关性,因此利用对历史气象探空数据及其仿真对流层延迟建立了不同高度的折射率与地面气象参数、对流层延迟之间的统计回归关系,由此建立了反演模型。4)通过引入利用神经网络方法、支持向量机方法等现代优化算法深入挖掘地面气象参数及对流层延迟与不同高度折射率的相关性关系,建立了多种反演模型和算法并进行对比研究,研究发现其中以基于不同高度角对流层延迟差的相关向量机反演方法的综合性能最优,该方法利用多种手段消除延迟测量误差对反演的影响,有效解决了反演方法对延迟测量的容错性问题。5)综合单站地基GNSS实时估计低高度角对流层斜延迟方法和基于低高度角斜延迟梯度的对流层折射率剖面相关向量机反演算法,提出了基于地基单站GNSS的对流层折射率剖面的实时反演方法,突破了实时反演的瓶颈,仿真和实验结果证明了该方法的有效性和优越性。本文为大气剖面参数的遥感反演提供了一种简便、高效、准确的新手段。本文的研究结果将有助于航天测控、导航定位、雷达探测等系统的大气的实时电波折射修正。更多还原

【Abstract】 The tropospheric refractivity profile is the basis for radio wave correction for refractive error, however, there is not a simple and convenient method to measure it for all-weather in real time at present. The Global Navigation Satellite System(GNSS) have been applied successfully not only in the traditional field such as ranging, time service, and navigation, but also in environmental monitoring, geodesy, astronomy, and meteorology and so on. On the basis of estimating electromagnetic wave refraction parameters (path delay and bending angle) by single ground-based GNSS receiver, the retrieving algorithms of tropospheric refractivity profile have been investigated herein. The main topics and results of the study are as follows:Firstly, the general methods of obtaining tropspheric delay by the Precise Point Position method (PPP) depend upon the precise IGS (the International GPS Service) final products, which are usually available on the thirteenth day after the last observation, therefore the tropospheric delays can not be acquired in real time. On the basis of the PPP method with IGS ultra-rapid products, new techniques are developed here of calclulating the benging angle and the slant path delay (STD) at the straight line elevation below 5°in real time. The experiment show that the RMS errors of the estimated STDs by the methods are less than 5%.Secondly, on the assumption that the tropospheric refractivity profile approximates to some model, the retrieving methods are proposed for estimating the parameters of the model by some searching algorithm by path delay of GNSS receiver. The main advantage of this method is that the historical radiosonde data are not necessary here.Thirdly, the statistical relationships between simulated path delay and atmospheric refractivity at various height based on historical radiosonde data have been analysed, and it is found that there are good correlations between them. Therefore the regressing models are founded between (a) surface meteorologic parameters and simulated tropospheric path delay and (b) tropospheric refractivity profile.Fourthly, the BP (Back Propagation) neutral network, SVM (Support Vector Machine), LS-SVM (Least-Squre SVM), and RVM (Relevance Vector Machine) are used to model tropospheric refactivity profile by the STDs, and these models are compared by simulation and experiment. The RVM model based on the differences of the STDs at various straight line elevations has the best performance, especially for tolerating the errors of the retrieved STDs.Fifthly, combining (a) the estimation method of the slant path tropospheric delay (STD) at low straight line elevation (<5°) of single ground-based GNSS receiver in real time and (b) the RVM inversion method based on the differences of STDs, a method is developed for retrieving atmospheric refractivity profiles in real time based on single ground-based GPS receiver, which solves the difficulty of inversion in real time. The simulation and experiment indicate the validity of the method.The studies in this dissertation propose a new method to retrieve atmospheric refractive profile in real time, at low cost, and high efficiency. The results herein will be helpful for correction for tropospheric refractive error of radio wave in electronic systems including TT&C (Telemetry, tracking, and commanding), navigation, radar, etc.更多还原