A GIS approach to GNSS simulation in urban canyons

GPS and other Global Navigation Satellite Systems (GNSS) (for example the Russian GLONASS system, the emerging European Galileo system and the Chinese COMPASS) are developing rapidly. In the future more than 100 satellites could be available for positioning. As a result, the ability to accurately position, within a level of accuracy of a few centimetres, in difficult areas, such as urban canyons, will be feasible. The current GNSS satellites (GPS and GLONASS) were tested along a trial route on the University of Nottingham Campus. Although the combination of current GNSS satellites increased the positioning availability compared with GPS alone, there are still large areas of degraded positional accuracy. Therefore, simulation for the forthcoming GNSS satellites in nearly similar environments will be necessary to test different GNSS combinations. In recent years, both GNSS and GIS have evolved as predominant mapping tools with significant potential for future development. A new GIS tool ‘Urban Canyon GNSS Simulation’ (UCGS) was developed by the authors to provide a powerful link between the two, using the powerful operations of GIS to optimize the capabilities of GNSS. This tool simulates satellite visibility and Dilution Of Precision (DOP) for both the current and the future GNSS constellations within a 3D city model environment. UCGS algorithms were tested against real data for GPS constellation that were collected during a period of 24 hours at a fixed point on the IESSG Tower building. The results show that 91.7% of the number of satellites match between the real and the simulated satellites. There were 473 of 5758 epochs with a difference of 1 satellite and 4 epochs with a difference of 2 satellites. Two simulation tests were carried out; one along a trial route at the University of Nottingham using a 3D TIN model created based on photogrammetry data. Another simulation was carried out at Upton-upon-Severn using a 3D model (raster model) created from LiDAR data. The results achieved from these tests, show that the number of satellites increased and DOP values dropped as more GNSS constellations were used in the simulation. For example one testing point in Nottingham’s test, surrounded by buildings with different heights, has 3 GPS satellites and about GDOP. The number of satellites increased to 5, 8, 11 and 15 and the GDOP reduced to about 45, 12, 8 and 4 after adding current GLONASS satellites, full GLONASS satellites, Galileo satellites and COMPASS satellites respectively. A full description about the UCGS tool as well as the results from different GNSS combinations is illustrated herein.