首页|基于图形处理器的水下目标传递函数多频点处理方法

基于图形处理器的水下目标传递函数多频点处理方法

扫码查看
为了提高水下目标宽带回波的计算速度,本文提出一种基于图形处理器GPU的散射传递函数多频点快速计算解决方案.相较于传统算法中逐个频率点计算的方式,CUDA快速算法充分利用各频点处目标强度的相对独立性,基于GPU的硬件特点,同时计算宽带内的散射声场,从而显著提高了计算效率.本文以潜航器模型为算例,对不同网格数量下模型的目标散射传递函数计算速度进行对比分析.仿真结果表明,相较于传统的CPU串行计算,采用CUDA快速算法能够实现超过80的加速比,有效提高了计算速度.
Deployment method of graphic processor for scattering transfer function of underwater targets
In order to improve the calculation speed of broadband echoes of underwater targets,this paper proposes a solution for fast computation of the scattering transfer function at multiple frequency points based on graphics processing unit(GPU).Compared to the traditional approach of calculating each frequency point individually,the CUDA fast algorithm fully utilizes the relative independence of target intensity at different frequency points and takes advantage of the hardware characteristics of GPUs to simultaneously compute the scattering sound field within the broadband,significantly improving the computational efficiency.The cylindrical model is used as a case study in this paper to compare and analyze the calcula-tion speed under different grid quantities.The simulation results demonstrate that,compared to traditional CPU serial compu-tation,utilizing the CUDA fast algorithm can achieve an acceleration ratio of over 80,effectively improving the calcula-tion speed.

planar elements methodgraphics processing unitcompute unified device architectureparallel com-puting

钱浩然、王斌

展开 >

上海交通大学船舶海洋与建筑工程学院,上海 200240

上海交通大学三亚崖州湾深海科技研究院,海南三亚 572024

板块元方法 图像处理器 计算统一设备架构 并行计算

三亚崖州湾科技城科技专项资助项目

SKJC-2022-PTDX-020

2024

舰船科学技术
中国舰船研究院,中国船舶信息中心

舰船科学技术

CSTPCD北大核心
影响因子:0.373
ISSN:1672-7649
年,卷(期):2024.46(14)
  • 4