Centre dedocumentation
scientifique

Automatic GPS ionospheric amplitude and phase scintillation detectors using a machine learning algorithm / Yu Jiao in Inside GNSS, vol 12 n° 3 (May - June 2017)  

Public [article]  inInside GNSS > vol 12 n° 3 (May - June 2017) . - pp 48 - 54 Titre : Automatic GPS ionospheric amplitude and phase scintillation detectors using a machine learning algorithm Type de document : Article/Communication Auteurs : Yu Jiao, Auteur ; John J. Hall, Auteur ; Yu T. Morton, Auteur Année de publication : 2017 Article en page(s) : pp 48 - 54 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement du signal [Termes IGN] amplitude [Termes IGN] apprentissage automatique [Termes IGN] détecteur [Termes IGN] propagation ionosphérique [Termes IGN] scintillation Résumé : (auteur) Ionospheric scintillation can cause errors or outage in GNSS services. Timely detection of ionospheric scintillation will enable adaptive processing to mitigate its effects on navigation solutions. This article presents a machine learning algorithm to autonomously detect ionospheric amplitude and phase scintillation. Systematic validations of the algorithm show the accuracies of amplitude and phase scintillation detections to be at around 98% and 92%, respectively. Tests were also conducted using data from locations not involved in training process and the results demonstrate the generalization capabilities of the scintillation detectors. In addition, concurrent amplitude and phase scintillation detection using similar machine learning algorithms is further investigated with low-latitude scintillation data. Our findings suggest that at low latitudes, amplitude detector alone is sufficient to capture scintillation in general. At high latitudes, phase scintillation is known to be the dominating phenomena and phase scintillation detector is necessary to capture the scintillation events. Numéro de notice : A2017-259 Affiliation des auteurs : non IGN Thématique : POSITIONNEMENT Nature : Article DOI : sans En ligne : http://www.insidegnss.com/node/5467 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=85276 [article]