Study on acceleration technique for calculating near field of horn antenna based on GPU
School of Science, Xidian University, Xi’an Shaanxi, China
IEEE International Conference on Microwave Technology & Computational Electromagnetics (ICMTCE), 2011
@inproceedings{su2011study,
title={Study on acceleration technique for calculating near field of horn antenna based on GPU},
author={Su, X. and Wu, Z. and Jiang, Z. and Wang, X.},
booktitle={Microwave Technology & Computational Electromagnetics (ICMTCE), 2011 IEEE International Conference on},
pages={470–473},
organization={IEEE},
year={2011}
}
Horn antennas are extremely popular in microwave region, so it has great practical significance of studying its near field. The radiation of horn antenna can be equivalent to that of the surface current on aperture by Huygens principle. Since the dipole is the simplest and most familiar antenna, we use the array of dipoles to replace the currents and get the same radiation effects. The compute unified device architecture (CUDA), which provides fine-grained data parallelism and thread parallelism, can accelerate the calculation of the radiation of the dipole array efficiently. At last a comparison of the parallel algorithm with the sequential algorithm is made to obtain the speedup.
July 10, 2011 by hgpu
