Title: SIAM: Extruded Shapefile based Interference Avoidance and Mitigation for GNSS Navigation in Urban Canyons
Author(s): Guoyu Fu, Colton Riedel, Tyler Holmes and Jyh-Charn Liu
Published in: Proceedings of the 30th International Technical Meeting of The Satellite Division of the Institute of Navigation (ION GNSS+ 2017)
September 25 - 29, 2017
Oregon Convention Center
Portland, Oregon
Pages: 3269 - 3284
Cite this article: Fu, Guoyu, Riedel, Colton, Holmes, Tyler, Liu, Jyh-Charn, "SIAM: Extruded Shapefile based Interference Avoidance and Mitigation for GNSS Navigation in Urban Canyons," Proceedings of the 30th International Technical Meeting of The Satellite Division of the Institute of Navigation (ION GNSS+ 2017), Portland, Oregon, September 2017, pp. 3269-3284.
Full Paper: ION Members/Non-Members: 1 Download Credit
Sign In
Abstract: This paper proposes an extruded -shapefile based interference avoidance and mitigation (SIAM) scheme for global navigation satellite system (GNSS) navigation in urban canyon environments. In SIAM, widely available metropolitan shapefiles are fused with digital elevation models (DEMs) produced via remote sensing technology such as LiDAR to generate an extrudedshapefile (e-shapefile) to represent both the footprints and heights of buildings. Using the e-shapefile and space vehicle (SV) positions reported in ephemeris data, the blockage or presence of line-of-sight (LOS) visibility for a position with respect to a given SV can be computed, aiding motorists and other GNSS users in differentiating between SVs in LOS or non-LOS (NLOS) positions. When four or more SVs have LOS to the receiver, the GNSS receiver can use standard navigation equations which operate with four degrees of freedom (DOF4) to resolve position and time. When less than four SVs are visible, we propose a horizon positioning algorithm with three degrees of freedom (DOF3) when the horizon planar transformation is combined with recent navigation history to obtain approximate position solutions. Experiments utilizing a software defined radio (USRP N200), a survey-grade GPS receiver (Trimble R10), and Brazos county, Texas LiDAR maps (6-inch resolution) were conducted on a stretch of the Texas A&M University (TAMU) campus, featuring both large structures and tree canopies. A LOS map was generated and the accuracy of DOF3 solutions excluding NLOS SVs were evaluated against DOF4 solutions obtained with a ublox receiver, as well as precision surveyed points from the Trimble R10, which was considered ground truth. Experimental results suggest that a LOS based SV selection scheme can correctly identify the LOS/NLOS status of SVs, and such selections can yield noticeable reductions in position errors due to multipath effects of surrounding buildings in a series of road tests. By combining the DOF4 and DOF3 solvers in a hybrid solver, our experiments show that it improved navigation results can be achieved in the presence of dense tree canopies and buildings.