Anxel Beam Shrinkage Method and Heterogeneous Computing-Accelerated Full-Image Theory Method Ray Tracing Enabling Massive Outdoor Propagation Modeling

Despite their accuracy, traditional image theory (IT) ray tracers were previously limited to basic simulation environments with fewer field observation points (FOPs) and lower ray bounce orders due to computational inefficiencies. In this study, we propose a novel full-3-D anxel beam shrinkage (ABS)...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on antennas and propagation Vol. 72; no. 7; pp. 5935 - 5949
Main Authors: Kim, Yongwan, Yang, Hyunjun, Kim, Hooyoung, Jo, Junpyo, Oh, Jungsuek
Format: Journal Article
Language:English
Published: New York IEEE 01.07.2024
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Subjects:
Accuracy
Algorithms
Asymptotic high-frequency techniques
Bouncing
Central Processing Unit
Central processing units
Computational modeling
CPUs
Diffraction
graphical processing unit (GPU)
Graphics processing units
image theory (IT) ray tracing (RT)
Parallel processing
Ray tracing
Reflection
shooting and bouncing ray (SBR) techniques
Shrinkage
Tracers
Visibility
wireless propagation modeling
ISSN:0018-926X, 1558-2221
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Despite their accuracy, traditional image theory (IT) ray tracers were previously limited to basic simulation environments with fewer field observation points (FOPs) and lower ray bounce orders due to computational inefficiencies. In this study, we propose a novel full-3-D anxel beam shrinkage (ABS) method and heterogeneous computing-accelerated full-IT method ray-tracing (RT) framework enabling massive outdoor propagation modeling. The proposed framework is divided into three components: 1) visibility preprocessing; 2) visibility tree generation, which introduces a novel ABS method to expedite the creation process and minimize the visibility tree's size; and 3) shadow testing and field calculation, incorporating a heterogeneous computing algorithm designed to efficiently manage numerous FOPs. We also demonstrated that the proposed framework, utilizing both central processing unit (CPU) and graphical processing unit (GPU) parallel computing, is 651 times faster than the IT method solver of WinProp, which supports only CPU parallel computing. Furthermore, it is confirmed that the proposed RT framework can handle <inline-formula> <tex-math notation="LaTeX">1\times 1 </tex-math></inline-formula> km wide and dense urban outdoor simulation with up to the maximum ray bouncing order of 6 and thousands of FOPs. The proposed RT framework could serve as a foundation for future advancements in IT method RT techniques in complex and massive scenarios, which were previously exclusive to the shooting and bouncing rays method ray tracers.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0018-926X
1558-2221
DOI:10.1109/TAP.2024.3411793