| 基于GPU 的液晶大气湍流模拟器的波面生成计算 |
| |
| 引用本文： | 孔悦,徐熙平,倪小龙.基于GPU 的液晶大气湍流模拟器的波面生成计算[J].红外与激光工程,2014,43(9):3061-3065. |
| |
| 作者姓名： | 孔悦 徐熙平 倪小龙 |
| |
| 作者单位： | 1.长春理工大学光电工程学院,吉林长春130022; |
| |
| 摘 要： | 提出了一种基于GPU 的液晶大气湍流模拟器实时波面生成的计算方法，为了让液晶空间光调制器进行大气湍流类比。依据液晶湍流模拟器高分辨率、高精度的特性讨论CUDA 的算法。此外，建立一种基于GPU 波面生成的模型并进一步对其优化。最后给出使用CPU 和GPU 后的结果并进行类比。结果表明:采用231 项Zernike 系数生成分辨率为256256 的波前所需时间少于2 ms，与传统的采用CPU 生成的方法相比速度提升两个量级，满足实时波面生成的要求。
|
| 关 键 词： | 图形处理器 CUDA 液晶 湍流计算 |
| 收稿时间： | 2014-01-05 |
Calculation of liquid crystal wave-front generation atmospheric turbulence simulator based on GPU |
| |
| Affiliation: | 1.School of Opto-Electronic Engineering,Changchun University of Science and Technology,Changchun 130022,China;2.Fundammontal Science on Space-Ground Laser Communication Technology Laboratory,Changchun University of Science and Technology,Changchun 130022,China |
| |
| Abstract: | In order to achieve the purpose of real-time atmosphere turbulence simulation by used liquid crystal spatial light modulator, A new computing method of liquid crystal atmospheric turbulence simulator real-time wave-front generation based on GPU has been proposed. According to the characteristics of liquid crystal turbulence simulator, which is high resolution, high precision, The Compute Unified Device Architecture was discussed in this paper. Furthermore, a wave-front generation model based on GPU was established with parallel optimization and share memory optimization. Finally, the experimental results by used CPU and GPU wave generation was given. The result shows that: the consumed time by proposed method is less than 2 ms for a Zernike polynomial with 231 wave-front values in resolution of 256256, which is two orders of magnitude less than that of CPU, fully meets the real-time wave-front generation requirements. |
| |
| Keywords: | |
| 本文献已被 CNKI 等数据库收录！ |
| 点击此处可从《红外与激光工程》浏览原始摘要信息 |
|
点击此处可从《红外与激光工程》下载全文 |
|
