부산대학교 도서관

Urban modeling is one of the most important parts of smart city development and requires the acquisition of urban geometric data, especially those regarding building boundaries. Traditional urban modeling methods rely on light detection and ranging (LiDAR), oblique photogrammetry, or mobile mapping systems to measure and rectify building geometrical parameters. However, these methods usually require costly devices and a lot of labor. This letter proposes a novel building model rectification method based on a consumer-grade GNSS receiver, which measures the perpendicular distance between building facades and a referencing location by GNSS reflectometry (GNSS-R) and raytracing. The performance of GNSS-R ranging is verified by three experiments with a total station and a LiDAR simultaneously. The results show that the proposed method enabled a smartphone with signal-to-noise ratio (SNR) observations to estimate building facade distances with a mean error of 5 cm. The preliminary results demonstrate the feasibility of this method to achieve precise urban modeling.