The acceleration of PO based on CUDA.

Autor: Su, Xiang, Wu, Zhensen, Lin, Jiaxuan, Wang, Xiaobing
Zdroj: 2012 IEEE International Conference on Cyber Technology in Automation, Control & Intelligent Systems (CYBER); 1/ 1/2012, p126-129, 4p
Abstrakt: The electromagnetic scattering of a complex target with large electrical size could not be solved by the analytical and the numerical methods. In this paper physical optics (PO) is used to solve this problem. Although compared with numerical method, such as the method of moment (MoM) and the finite difference time domain (FDTD), the PO costs shorter executing time and needs less memory, it should judge the patches whether they are illuminated or not before calculating the induced current. This preprocess often needs a long time. The graphic process unit (GPU) based on compute unified device architecture (CUDA) is used to solve the problem parallel. [ABSTRACT FROM PUBLISHER]
Databáze: Complementary Index