Multivariate analysis of combined GPS/GLONASS point positioning performance in Brazilian regions under different ionospheric conditions

Abstract

The development of Global Navigation Satellite Systems (GNSS) was a revolution in activities related to positioning. Currently, GLONASS (Global'naya Navigatsionnaya Sputnikovaya Sistema) and GPS (Global Positioning System) are the main systems with full constellation. The use of GPS and GLONASS combined data gained renewed attention after the GLONASS restoration and modernization plan, which enabled the system to reach full constellation in 2011. In addition to the use of combined data, several other factors can influence positioning quality, such as the methods applied and errors that can affect the transmitted signals. Concerning errors, the ionosphere is an important source, particularly for users of single frequency receivers. This requires special attention, because, in addition to degradation of positioning accuracy, there is a great interdependency between signal loss and ionospheric irregularities, such as ionospheric scintillation. In this paper, multivariate analysis techniques were applied to investigate the influence of ionospheric activity, specifically ionospheric scintillation, in positioning error. An experiment was carried out applying the point positioning method considering stations located in different places during periods of high and low ionospheric activity. The results showed high similarity between GPS and GLONASS data and significant ionosphere influence in the positioning error. S4 indexes presented correlations higher than 0.75 when considering GPS and GLONASS data. Positioning error using GPS or GPS/GLONASS data presented correlations higher than 0.93 for all stations considered. Techniques such as clustering, correspondence analysis and factor analysis were also applied in this study.