3D Line-of-Sight Analysis of Wireless Signal Coverage in the United States
Topics:
Keywords: GIS, Spatial database, Line-of-Sight, Viewshed analysis, Wireless signals
Abstract Type: Virtual Paper Abstract
Authors:
Xin Yu, Spatial Science Institute, University of Southern California, Los Angeles, CA, USA
Bassel Abou Ali Modad, Department of Electrical and Computer Engineering, University of Southern California, Los Angeles, CA, USA
I-Hui Huang, Computer Science Department, University of Southern California, Los Angeles, CA, USA
Sudan Zhang, Computer Science Department, University of Southern California, Los Angeles, CA, USA
Hae Jin Song, Department of Computer Science and Information Sciences Institute, University of Southern California, Marina del Rey, CA, USA
Andreas F. Molisch, Department of Electrical and Computer Engineering, University of Southern California, Los Angeles, CA, USA
Yao-Yi Chiang, Department of Computer Science and Engineering, University of Minnesota, Minneapolis, MN, USA
,
,
,
Abstract
Viewshed analysis from cell towers is essential for calculating the coverage of wireless signals and planning for cell tower placement. Existing methods cannot handle computationally intensive viewshed calculation in large areas with complex landscapes. This study proposes a parallel raster-vector mixed LOS calculation method incorporating the digital elevation model (DEM) and building features from OpenStreetMap (OSM) to generate the LOS propagation from cell towers on primary roads in the continental US using a spatial database, PostgreSQL and PostGIS. We define the cell towers, collected from Federal Communications Commission Antenna Structure Registration, as observer points and 10m evenly sampled road points within 1 km buffers of cell towers as target points. The visibility of each target point is calculated by comparing the hypothetical height of 10m-sample points on the 3D line between itself and the cell tower with the actual total height from the DEM and OSM building height. A target point is visible from the cell tower only if all the sample points on the hypothetical 3D line are not blocked. We divided the continental US into 491 level-3 geohashing cells (about 156*156 km2) to conduct parallel processing. We assigned the nearest centroids of level-9 geohashing cells (about 4.77*4.77 m2) as approximated locations of line points on hypothetical 3D lines to accelerate queries of real-world elevation and building height by reducing the repetitions. For 161,127 towers in the continental US, 87.7% road points are visible from cell towers within 1 km (84.7% for urban and 93.3% for rural areas).
3D Line-of-Sight Analysis of Wireless Signal Coverage in the United States
Category
Virtual Paper Abstract
