The use of Global Navigation Satellite System (GNSS) for localisation purposes needs previous data analysis and an evaluation of the generated position information based on an independent reference, in order to validate the whole localisation system. An accuracy analysis based on Mahalanobis Ellipses Filter (MEF) is presented. This new kind of filtering allows a better comprehension of the nature of the accuracy (trueness and precision) of the deviation dataset; and a better ground for GNSS receivers' validation. MEFs methodology focuses not only in the finding of outliers from the dataset, but also in the meaning of the resulting Mahalanobis ellipses evaluation in relation to localisation tasks. The Mahalanobis distance is a distance measure based on the correlation between variables. This distance allows identifying different patterns within related datasets, by analyzing the similarity between unknown and known samples. The difference between Euclidean distance and Mahalanobis distance is that the last one takes into account the correlation of the entire dataset, resulting in a scale-invariant distance; a multivariate effect size. This kind of distance has been used for several purposes in the last years, but it has never been displayed as ellipses for accuracy description of intelligent transportation system (ITS). The presented evaluation by means of MEFs gives a better description of the accuracy characteristic of a GNSS-based system and the GNSS receiver. This new filtering method will be very useful for future developments of GNSS validation tools and GNSS receivers' certification methodologies.


    Zugriff

    Zugriff über TIB

    Verfügbarkeit in meiner Bibliothek prüfen


    Exportieren, teilen und zitieren



    Titel :

    New filter by means of Mahalanobis distance for accuracy evaluation of GNSS




    Erscheinungsdatum :

    2013


    Format / Umfang :

    9 Seiten, Bilder, Tabellen, 7 Quellen


    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Datenträger


    Sprache :

    Englisch




    An adaptive fading Kalman filter based on Mahalanobis distance

    Chang, Guobin / Liu, Ming | SAGE Publications | 2015



    Directional Mahalanobis Distance and Parameter Sensitivities

    Chinta, Balakrishna | SAE Technical Papers | 2016


    Vector Quantization Based on Dynamic Adjustment of Mahalanobis Distance

    Younis, K. S. / Rogers, S. K. / DeSimio, M. P. et al. | British Library Conference Proceedings | 1996


    Local Binary Patterns with Mahalanobis Distance Maps for Human Detection

    He, Xiangjian / Li, Jianmin / Chen, Yan | IEEE | 2008