The aim of this paper is to propose a multipath detection algorithm, which is based on K-means clustering that belongs to the class of unsupervised machine learning algorithms. The algorithm processes measurement sets computed for each satellite, namely, carrier phase, pseudorange and carrier-to-noise ratio, creating clusters of consistent measurements, thus allowing the identification of satellite signals suffering from the multipath error. Since it is an unsupervised method, it overcomes one of the most limiting features of supervised algorithms that require training data sets a-priori obtained as representative of multipath and no-multipath conditions. The study exploits both the real GNSS data affected by the multipath in the surrounding environment that were collected at South African Antarctic research base SANAE-IV and the simulated data where the ionospheric, tropospheric and multipath errors are modelled. Receiver Autonomous Integrity Monitoring (RAIM) algorithm with parity method was also implemented and tested for the same datasets, and it will be used as a term of comparison for the algorithm performance.

Multipath detection based on K-means clustering / Savas, C.; Dovis, F.. - ELETTRONICO. - (2019), pp. 3801-3811. ((Intervento presentato al convegno 32nd International Technical Meeting of the Satellite Division of the Institute of Navigation, ION GNSS+ 2019 tenutosi a Hyatt Regency Miami, USA nel 2019 [10.33012/2019.17028].

Multipath detection based on K-means clustering

Savas C.;Dovis F.
2019

Abstract

The aim of this paper is to propose a multipath detection algorithm, which is based on K-means clustering that belongs to the class of unsupervised machine learning algorithms. The algorithm processes measurement sets computed for each satellite, namely, carrier phase, pseudorange and carrier-to-noise ratio, creating clusters of consistent measurements, thus allowing the identification of satellite signals suffering from the multipath error. Since it is an unsupervised method, it overcomes one of the most limiting features of supervised algorithms that require training data sets a-priori obtained as representative of multipath and no-multipath conditions. The study exploits both the real GNSS data affected by the multipath in the surrounding environment that were collected at South African Antarctic research base SANAE-IV and the simulated data where the ionospheric, tropospheric and multipath errors are modelled. Receiver Autonomous Integrity Monitoring (RAIM) algorithm with parity method was also implemented and tested for the same datasets, and it will be used as a term of comparison for the algorithm performance.
0-936406-23-2
978-093640623-7
File in questo prodotto:
File Dimensione Formato  
IONGNSS19-0252.pdf

non disponibili

Descrizione: Online proceedings version
Tipologia: 2a Post-print versione editoriale / Version of Record
Licenza: Non Pubblico - Accesso privato/ristretto
Dimensione 1.27 MB
Formato Adobe PDF
1.27 MB Adobe PDF   Visualizza/Apri   Richiedi una copia
Pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11583/2772092
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo