Share Email Print
cover

Proceedings Paper

GPU-based ray tracing algorithm for high-speed propagation prediction in typical indoor environments
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

A fast 3-D ray tracing propagation prediction model based on virtual source tree is presented in this paper, whose theoretical foundations are geometrical optics(GO) and the uniform theory of diffraction(UTD). In terms of typical single room indoor scene, taking the geometrical and electromagnetic information into account, some acceleration techniques are adopted to raise the efficiency of the ray tracing algorithm. The simulation results indicate that the runtime of the ray tracing algorithm will sharply increase when the number of the objects in the single room is large enough. Therefore, GPU acceleration technology is used to solve that problem. As is known to all, GPU is good at calculation operation rather than logical judgment, so that tens of thousands of threads in CUDA programs are able to calculate at the same time, in order to achieve massively parallel acceleration. Finally, a typical single room with several objects is simulated by using the serial ray tracing algorithm and the parallel one respectively. It can be found easily from the results that compared with the serial algorithm, the GPU-based one can achieve greater efficiency.

Paper Details

Date Published: 20 October 2015
PDF: 11 pages
Proc. SPIE 9646, High-Performance Computing in Remote Sensing V, 964606 (20 October 2015); doi: 10.1117/12.2197384
Show Author Affiliations
Lixin Guo, Xidian Univ. (China)
Xiaowei Guan, Xidian Univ. (China)
Zhongyu Liu, Xidian Univ. (China)


Published in SPIE Proceedings Vol. 9646:
High-Performance Computing in Remote Sensing V
Bormin Huang; Sebastián López; Zhensen Wu; Jose M. Nascimento; Boris A. Alpatov; Jordi Portell de Mora, Editor(s)

© SPIE. Terms of Use
Back to Top