Catalogue en ligne IGN

Fast precise GPS positioning in the presence of ionospheric delays / Dennis Odijk (2002)

Public

Titre :	Fast precise GPS positioning in the presence of ionospheric delays
Type de document :	Thèse/HDR
Auteurs :	Dennis Odijk, Auteur
Editeur :	Delft : Netherlands Geodetic Commission NGC
Année de publication :	2002
Collection :	Netherlands Geodetic Commission Publications on Geodesy, ISSN 0165-1706 num. 52
Importance :	242 p.
Format :	16 x 24 cm
ISBN/ISSN/EAN :	978-90-6132-278-8
Note générale :	Bibliographie
Langues :	Anglais (eng)
Descripteur :	[Vedettes matières IGN] Géodésie spatiale [Termes IGN] ambiguïté entière [Termes IGN] correction ionosphérique [Termes IGN] correction troposphérique [Termes IGN] données GPS [Termes IGN] interpolation [Termes IGN] krigeage [Termes IGN] mesurage de phase [Termes IGN] méthode des moindres carrés [Termes IGN] modèle de Gauss-Markov [Termes IGN] modèle ionosphérique [Termes IGN] modèle stochastique [Termes IGN] propagation ionosphérique [Termes IGN] propagation troposphérique [Termes IGN] réfraction atmosphérique [Termes IGN] résolution d'ambiguïté [Termes IGN] signal GPS [Termes IGN] station virtuelle [Termes IGN] traitement de données GNSS [Termes IGN] traitement du signal
Index. décimale :	30.61 Systèmes de Positionnement par Satellites du GNSS
Résumé :	(Auteur) This thesis deals about geodetic applications of the Global Positioning System (GPS), in which the position of the GPS receiver must be determined with cm precision. This requires a relative measurement setup, together with an advanced processing strategy based on observations of the carrierphase of the signal. To keep it economically interesting, this CPS technique should be based on relatively short time spans in which the satellite observations are collected. The key to precise positioning using short time spans is to take advantage of the integer property of the ambiguities of the phase observations in the processing. The above procedure has been applied in a successful way for the last decade to applications in which the distance between the receivers is restricted to about 10 km (the socalled rapidstatic and realtime kinematic GPS techniques over short distances). Above this distance, it is known that certain errors in the GPS observations start to significantly bias the computed receiver position when they are not taken care of. The aim of this research therefore is to develop a processing procedure, taking into account the errors in GPS observations due to propagation of the signals through the ionosphere, the atmospheric layer above about 80 kill. Although other errors (due to troposphere and satellite orbit) are of relevance as well, the research is restricted to an improved modelling of the ionospheric error. since it is by far the largest error. For the other errors standard modelling techniques are applied in this research. Using the procedure, it should be possible to determine the desired receiver positions with cmprecision using a short tinle span. The research is restricted to GPS receivers with a mutual distance of a few hundred km (mediumdistance baselines), located in midlatitude regions. To facilitate a modelling of the ionospheric error, using the theor ' y of atmospheric refraction it is possible to decompose this error into a firstorder effect, which contains the gross of the error, plus some higherorder effects and a term due to bending of the signal path. Under worstcase conditions. the firstorder term may range up to about 80 m (on the GPS L2 frequency), whereas the accumulated effect of higherorder and bending terms can be tip to 4 cm (for L2). For the future L5 frequency (from 2008) these effects are even larger. Fortunately, because of the relative setup and the assumed medium distances, it is proved for this research it is allowed to neglect the higherorder and bending errors. In the procedure a stochastic modelling of the firstorder ionospheric errors (referred to as ionospheric delays) is chosen. This means that the ionospheric delays are not modelled as completely unknown parameters, but that stochastic prior information is incorporated by means of ionospheric pseudoobservations. This model is referred to as the ionosphereweighted model: The weight of the ionospheric information can be tuned by the a priori standard deviation of the pseudoobservations. When this standard deviation is chosen zero, the ionosphereweighted model reduces to the ionospherefixed model, which is the usual processing model for shortdistance baselines (for which the ionospheric delays may be neglected). On the other hand, with an infinitely large ionospheric standard deviation, the model will be equivalent to the ionospherefloat model, in which the ionospheric delays are assumed as completely unknown parameters. This latter model is closely related to the ionospherefree combination, for which it is known that it cannot be used to achieve fast positioning results. It is shown that the ionosphereweighted model is only suitable for fast ambiguity resolution (and consequently positioning), when the ionospheric standard deviation is small. This requires very precise a priori ionospheric information. The developed procedure consists of three steps. It is required that a user collects CPS observations in the vicinity of a network of permanent GPS stations. In the first step, the observations at the network stations are processed simultaneously using the ionosphereweighted model. Since in this research the goal is precise positioning within the shortest time span possible, i.e. instantaneous or singleepoch positioning, it is required that the network data is also processed instantaneously. To make instantaneous resolution of the network ambiguities possible, the sample values of the ionospheric pseudoobservations are temporal predictions based on estimates of previous epochs. Test computations using a network with a station spacing of more than 100 km demonstrated that in this way high network ambiguity success rates (close to 100%) can be obtained. In the second step, precise ambiguityfixed network ionospheric delays are spatially interpolated at the approximate location of the user's receiver. In the procedure for this purpose the concept of virtual reference station (VRS) observations is used. In this concept the network estimates (ionospheric delays and other parameters) are transformed to VRS observations. which should correspond to the data a real receiver would have collected at the user's location. The processing of the user's observations relative to this VRS is the third step of the procedure. Because of the presence of possible residual ionospheric delays also in this step the ionosphereweighted model is applied. The difference with the application in the network processing is that the sample values of the pseudoobservations are now taken zero. and the ionospheric standard deviation is computed as a function of the distance to the closest real network station. Using this, test computations demonstrated that instantaneous ambiguity success rates of 90% are feasible. When the ionospherefixed model would be applied, the success rates would not be higher than about 60%.
Numéro de notice :	13101
Affiliation des auteurs :	non IGN
Thématique :	POSITIONNEMENT
Nature :	Thèse étrangère
DOI :	sans
En ligne :	https://www.ncgeo.nl/downloads/52Odijk.pdf
Format de la ressource électronique :	URL
Permalink :	https://documentation.ensg.eu/index.php?lvl=notice_display&id=54884

Exemplaires(1)

Code-barres	Cote	Support	Localisation	Section	Disponibilité
13101-01	30.61	Livre	Centre de documentation	Géodésie	Disponible

IGN

Centre de documentation
scientifique

Accueil

Sélection de la langue

Adresse

Se connecter

Actualité

L'actu ! Horaires du CDOS

Informations pratiques

Exemplaires(1)

IGN / ENSG

L'IGN a pour vocation

Accès directs

2014 -2024 IGN

IGN

Centre de documentationscientifique

Accueil

Sélection de la langue

Adresse

Se connecter

Actualité

L'actu ! Horaires du CDOS

Informations pratiques

Exemplaires(1)

IGN / ENSG

L'IGN a pour vocation

Accès directs

2014 -2024 IGN

Centre de documentation
scientifique