期刊文献+

基于GPU的电磁计算与SAR回波仿真加速方法研究

Research on Acceleration Method of GPU-based Electromagnetic Calculation and SAR Echo Simulation
下载PDF
导出
摘要 电磁计算与SAR(Synthetic aperture Radar,合成孔径雷达)回波仿真算法在雷达系统设计、雷达算法验证与目标自动识别算法研究等方面起着重要的作用,针对传统的电磁计算与SAR回波仿真算法耗时过长的问题,本文提出了一种基于GPU(Graphic Processing Unit,图形处理器)的电磁计算与SAR回波仿真全流程加速方法。首先,将传统算法中的三个核心计算部分——射线跟踪、电磁计算与SAR回波仿真进行多线程划分,然后利用GPU通用计算平台,并行处理所有线程单元,加快仿真速度。最后,实验结果表明:该方法相对于单核CPU计算的加速比达到450左右,相对于10核心CPU计算的加速比达到50左右。该方法在保证生成数据与传统算法一致的前提下,实现了良好的加速效果。 Electromagnetic calculation and SAR echo simulation algorithm play an important role in radar system design,radar algorithm verification,and automatic target recognition algorithms research.To solve the problem of traditional electromagnetic calculation and SAR echo simulation algorithm being too time-consuming,an acceleration method for the entire process based on GPU is proposed.Firstly,the three core computing parts of the traditional algorithm-ray tracing,electromagnetic calculation,and SAR echo simulation are divided into multiple threads.Then,by using the GPU general computing platform,all thread units are processed in parallel to accelerate the simulation speed.Finally,the experimental results show that the acceleration ratio of this method is about 450 compared with single-core CPU and about 50 compared to a 10-core CPU.This method achieves a good acceleration effect on the premise that the generated data is consistent with the traditional algorithm.
作者 吴雪 夏伟杰 蔡炜锋 WU Xue;XIA Weijie;CAI Weifeng(College of Electronic Information Engineering,Nanjing University of Aeronautics and Astronautics,Nanjing 210016,China)
出处 《遥测遥控》 2024年第4期108-114,共7页 Journal of Telemetry,Tracking and Command
基金 航天科技集团第一研究院复杂电磁环境下海面舰船回波仿真(1004-KFA21530)。
关键词 射线跟踪 电磁计算 SAR回波 全流程加速 并行计算 Ray tracing Electromagnetic computing SAR echo simulation Full process acceleration Parallel computing
  • 相关文献

参考文献14

二级参考文献50

  • 1林愿,吴淑泉,姜发玉.基于PCI接口的合成孔径雷达目标模拟器的设计与实现[J].测控技术,2005,24(3):42-44. 被引量:4
  • 2唐禹,王岩飞,张冰尘.滑动聚束SAR成像模式研究[J].电子与信息学报,2007,29(1):26-29. 被引量:29
  • 3Cumming I G and Wong F H. Digital Processing of SyntheticAperture Radar Data: Algorithms and Implementation [M].Norwood: Artech House, 2002. 被引量:1
  • 4OpenMP Architecture Review Board. OpenMP applicationprogram interface [OL]. http://www.openmp.CMrg/mp-documents/spec30.pdf, May 2008. 被引量:1
  • 5Snir M, Otto S, Lederman S H, et al" MPI: The CompleteReference[M]. US: The MIT Press, 1996. 被引量:1
  • 6NVIDIA. CUD A Cprogramming guide[OL]. http://docs.nvidia.com/cuda/pdf/CUD A _C_ Programming_Guide.pdf,July 2013. 被引量:1
  • 7NVIDIA. Tesla k20 GPU active accelerator: boardspecification [OL]. http://www.nvidia.cn/content/PDF/kepler/tesla-k20-active-bd-06499-001-v03.pdf. May 2013. 被引量:1
  • 8Intel. Intel Xeon Processor E5645 [OL]. http://ark.mtel.com/zh-cn/products/48768. 2010. 被引量:1
  • 9Yegulalp A F. Fast backprojection algorithm for syntheticaperture radar[C]. Proceedings 1999 IEEE Radar Conference,Waltham, US, 1999: 60-65. 被引量:1
  • 10John Jr. C K. Motion compensation for synthetic apertureradar [J]. IEEE Transaction on Aerospace and ElectronicSystems, 1975,11(3): 338-348. 被引量:1

共引文献25

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部