Centre de documentation
scientifique

Analyse des stochastischen Modells von GPS-Trägerphasenbeobachtungen / J. Howind (2005)

Public Titre : Analyse des stochastischen Modells von GPS-Trägerphasenbeobachtungen Titre original : [Analyse de modèles stochastiques des observations de phase GPS] Type de document : Thèse/HDR Auteurs : J. Howind, Auteur Editeur : Munich : Bayerische Akademie der Wissenschaften Année de publication : 2005 Collection : DGK - C Sous-collection : Dissertationen num. 584 Importance : 103 p. Format : 21 x 30 cm ISBN/ISSN/EAN : 978-3-7696-5023-5 Note générale : Bibliographie Langues : Allemand (ger) Descripteur : [Vedettes matières IGN] Géodésie spatiale [Termes IGN] base géodésique [Termes IGN] covariance [Termes IGN] données GPS [Termes IGN] double différence [Termes IGN] hauteur ellipsoïdale [Termes IGN] matrice de covariance [Termes IGN] méthode des moindres carrés [Termes IGN] modèle stochastique [Termes IGN] phase GPS [Termes IGN] précision du positionnement [Termes IGN] résidu [Termes IGN] série temporelle [Termes IGN] traitement de données GNSS Index. décimale : 30.61 Systèmes de Positionnement par Satellites du GNSS Résumé : (Auteur) Geodetic observations are usually processed by means of the Least Squares Method. However, this algorithm delivers reliable valuations of unknown parameters und associated accuracy measures only if the functional and stochastic models are appropriate. Observations using the Global Positioning System (GPS) are affected by various influences, e.g. by the atmosphere. Hence, it seems to be a reasonable conclusion that the commonly used stochastic model of GPS-observations with a scaled identity matrix with homoscedastic (i.e. equal variance) and uncorrelated errors is not adequate. The knowledge of the stochastic behaviour of GPS-observations can be improved by analyzing time series of residuals originating from Least Squares evaluation. Based on this analysis a procedure for modifying the covariance matrix of double-differenced GPS carrier-phase observations is presented. Essentially, it consists of two steps. At first individual variance functions for all double-difference observations depending on the satellite elevation for modifying the main diagonal of its covariance matrix are estimated. Secondly, temporal correlations are determined to adapt the other parts of the covariance matrix. All steps are verified by suitable statistical tests. The procedure is applied to extensive data sets with different characteristics. The results of the evaluation process, taking into account all realizations of the stochastic model mentioned above, are presented. It is shown that modifying the main diagonal of the covariance matrix with elevation-dependent variance functions leads to significant changes in the estimated ellipsoidal height of GPS-stations only for long baselines. The accuracy measures are not affected by this modification. In contrast to these results, considering temporal correlations in the covariance matrix of GPS-double difference observations causes distinct changes in the accuracy measures of all coordinate components independently of the baseline length. Numéro de notice : 13272 Affiliation des auteurs : non IGN Thématique : POSITIONNEMENT Nature : Thèse étrangère Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=54951

Exemplaires(2)

Code-barres	Cote	Support	Localisation	Section	Disponibilité
13272-01	30.61	Livre	Centre de documentation	Géodésie	Disponible
13272-02	30.61	Livre	Centre de documentation	Géodésie	Disponible