基于冗余计算约简的环扫SAR回波多GPU快速模拟

胡辰; 张帆; 李国君; 李伟; 崔忠马

doi:10.12000/JR15078

基于冗余计算约简的环扫SAR回波多GPU快速模拟

DOI: 10.12000/JR15078

1.
(北京化工大学信息科学与技术学院北京 100029)
2.
(北京遥感设备研究所北京 100854)

基金项目:

国家自然科学基金(61501018, 61302164),中央高校基本科研业务费专项资金(YS1404),北京高等学校青年英才计划(YETP0500)

详细信息

作者简介:
胡辰(1991-),男,山东人,2013年获得北京化工大学通信工程专业学士学位,同年保送至本校计算机技术专业攻读硕士学位。研究方向为SAR信号处理、高性能计算技术。E-mail:huchen_buct@163.com;张帆(1981-),男,2008年毕业于中科院电子所信号与信息处理专业,获得工学博士学位,同年在中科院电子所从事博士后研究,现为北京化工大学副教授,硕士生导师。主要研究方向为SAR系统模拟、高性能计算、科学可视化等。E-mail:zhangf@mail.buct.edu.cn;李国君(1991-),女,山西人,2013年获得北京化工大学电子信息工程专业学士学位,同年保送至本校计算机科学与技术专业攻读硕士学位。研究方向为SAR成像处理、高性能计算技术。E-mail:13269198197@163.com;李伟(1985-),男,2012年毕业于美国密西西比州立大学电气与计算机工程专业,获得工学博士学位,同年至美国加州大学戴维斯分校从事博士后研究,现为北京化工大学教授,博士生导师。主要研究方向为模式识别、高光谱图像分析与应用、数据压缩等。E-mail:liw@mail.buct.edu.cn;崔忠马(1978-),男,北京遥感设备研究所研究员,研究方向为雷达系统设计。

通讯作者:
张帆zhangf@mail.buct.edu.cn

计量
- 文章访问数: 3439
- HTML全文浏览量: 400
- PDF下载量: 1282
- 被引次数: 0
出版历程
- 收稿日期: 2015-06-18
- 修回日期: 2015-09-14
- 网络出版日期: 2016-08-28

Computation Reduction Oriented Circular Scanning SAR Raw Data Simulation on Multi-GPUs

1.
(College of Information Science and Technology, Beijing University of Chemical Technology, Beijing 100029, China)
2.
(Beijing Remote Sensing Equipment Research Institute, Beijing 100854, China)

Funds:

The National Natural Science Foundation of China (61501018, 61302164), The Fundamental Research Funds for the Central Universities (YS1404), The Beijing Higher Education Young Elite Teacher Project (YETP0500)

摘要

摘要: 环扫SAR作为一种特殊工作模式雷达,在对地观测方面有着广泛应用。随着分辨率提高及测绘带宽增大,对环扫SAR成像精度提出了更高的要求。而快速的海量回波模拟方法能够对高精度成像算法设计、研究提供有力的支撑。本文给出一种基于多GPU(Graphics Processing Unit,图形处理器)的环扫SAR回波模拟方法,并在此基础上进行冗余计算优化,通过MPI(Message Passing Interface,消息传递接口)在多GPU上进行了实现,实验结果表明在使用4块GPU的条件下,经过冗余计算约简,并行效率提高2倍以上,硬件成本降低50%,相对传统CPU串行仿真提高350倍左右。
- 环扫SAR /
- 回波模拟 /
- GPU并行 /
- 计算优化
Abstract: As a special working mode, the circular scanning Synthetic Aperture Radar (SAR) is widely used in the earth observation. With the increase of resolution and swath width, the simulation data has a massive increase, which boosts the new requirements of efficiency. Through analyzing the redundancy in the raw data simulation based on Graphics Processing Unit (GPU), a fast simulation method considering reduction of redundant computation is realized by the multi-GPUs and Message Passing Interface (MPI). The results show that the efficiency of 4-GPUs increases 2 times through the redundant reduction, and the hardware cost decreases by 50%, thus the overall speedup achieves 350 times than the traditional CPU simulation.
- Circular scanning SAR /
- Raw data simulation /
- GPU parallel /
- Computing optimization

HTML全文

参考文献(20)

[1]	李天池, 周荫清, 蔡世学, 等. 机载环扫成像雷达数据处理方法研究[J]. 系统工程与电子技术, 2002, 24(7): 1-3.Li Tianchi, Zhou Yinqing, Cai Shixue, et al.. Study on the data processing methods for airborne circular scanning imaging radar[J]. Systems Engineering and Electronics, 2002, 24(7): 1-3.
[2]	李静, 黄培康, 潘旭东, 等. 雷达下视环扫成像分辨率研究[J]. 系统工程与电子技术, 2009, 31(2): 315-318.Li Jing, Huang Peikang, Pan Xudong, et al.. Study on resolution for circular scanning imaging radar[J]. Systems Engineering and Electronics, 2009, 31(2): 315-318.
[3]	李勇, 朱岱寅, 朱兆达. 环视 SAR 成像处理中的几何失真校正算法[J]. 南京航空航天大学学报, 2009, 41(2): 232-237.Li Yong, Zhu Daiyin, and Zhu Zhaoda. Geometric distortion correction algorithm for circular-scanning SAR imaging[J]. Journal of Nanjing University of Aeronautics Astronautics, 2009, 41(2): 232-237.
[4]	毛新华, 朱岱寅, 李勇, 等. 环视 SAR 几何失真校正误差分析及补偿技术研究[J]. 电子与信息学报, 2007, 30(11): 2706-2709.doi:10.3724/SP.J.1146.2007.00679.Mao Xinhua, Zhu Daiyin, Li Yong, et al.. Study on error analysis of geometric distortion correction and compensation techniques for circular-scan SAR[J]. Journal of Electronics Information Technology, 2007, 30(11): 2706-2709. doi:10.3724/SP.J.1146.2007.00679.
[5]	赵亮, 毛新华, 吴迪. 基于回波信号的环视 SAR 成像运动参数估计[J]. 数据采集与处理, 2014, 29(4): 590-596.Zhao Liang, Mao Xinhua, and Wu Di. Motion parameter estimate of circular-scanning SAR imaging based on echo signal[J]. Journal of Data Acquisition and Processing, 2014, 29(4): 590-596.
[6]	孙兵, 周荫清, 李天池, 等. 环扫 SAR 的快速聚焦成像算法[J]. 北京航空航天大学学报, 2007, 33(7): 803-806.Sun Bing, Zhou Yinqing, Li Tianchi, et al.. Fast focused imaging algorithm for circular scanning SAR[J]. Journal of Beijing University of Aeronautics and Astronautics, 2007, 33(7): 803-806.
[7]	苏宇, 齐向阳. 基于OpenMP的星载SAR回波信号并行仿真[J]. 中国科学院研究生院学报, 2008, 25(1): 129-135.Su Yu and Qi Xiangyang. OpenMP based space-borne SAR raw signal parallel simulation[J]. Journal of Graduate School of the Chinese Academy of Sciences, 2008, 25(1): 129-135.
[8]	王曦爽, 黄立胜, 王贞松. 分布式星载SAR回波仿真的并行化计算研究[J]. 系统仿真学报, 2006, 18(8): 2097-2100.Wang Xishuang, Huang Lisheng, and Wang Zhensong. Research on parallel arithmetic of distribute spaceborne SAR ground target simulation[J]. Journal of System Simulation, 2006, 18(8): 2097-2100.
[9]	张帆, 林殷, 洪文. 基于网格计算的SAR回波分布式仿真[J]. 系统仿真学报, 2008, 20(12): 3165-3167.Zhang Fan, Lin Yin, and Hong Wen. SAR echo distributed simulation based on grid computing[J]. Journal of System Simulation, 2008, 20(12): 3165-3167.
[10]	Yu L, Xie X, and Xiao L. GPU-accelerated circular SAR echo data simulation of large scenes[C]. XXXIth URSI General Assembly and Scientific Symposium, Beijing, China, 2014: 1-4.
[11]	Chapman W, Ranka S, Sahni S, et al.. Parallel processing techniques for the processing of synthetic aperture radar data on GPUs[C]. IEEE International Symposium on Signal Processing and Information Technology, 2011: 573-580.
[12]	Zhu H, Xu H, and Feng L. Application of GPU for missile-borne SAR raw signal simulation[C]. International Conference on Artificial Intelligence, Management Science Electronic Commerce, Zhengzhou, China, 2011: 2816-2820.
[13]	Sheng H, Zhou M, Wang K, et al.. SAR echo simulation from numerous scattering cells based on GPU[C]. IET International Radar Conference, 2013: 1-5.
[14]	秦洁, 张志敏. 基于图形处理单元架构的合成孔径雷达回波仿真实现与优化[J]. 科学技术与工程, 2014, 14(13): 85-89.Qin Jie and Zhang Zhimin. Implementation and optimization of SAR echo simulation based on GPU[J]. Science Technology and Engineering, 2014, 14(13): 85-89.
[15]	Wang B, Zhang F, and Xiang M. SAR raw signal simulation based on GPU parallel computation[C]. IEEE Geoscience and Remote Sensing Symposium, Cape Town, South Africa, 2009: 617-620.
[16]	Zhang F, Wang B, and Xiang M. Accelerating InSAR raw data simulation on GPU using CUDA[C]. IEEE Geoscience and Remote Sensing Symposium, Honolulu, Hawaii, USA, 2010: 2932-2935.
[17]	Zhang F, Li Z, Wang B, et al.. Hybrid general-purpose computation on GPU (GPGPU) and computer graphics synthetic aperture radar simulation for complex scenes[J]. International Journal of Physical Sciences, 2012, 7(8): 1224-1234.
[18]	Zhang F, Hu C, Li W, et al.. Accelerating time-domain SAR raw data simulation for large areas using multi-GPUs[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2014, 7(9): 3956-3966.
[19]	黄立胜, 王贞松, 郑天垚. 基于FFT的快速SAR分布目标回波模拟算法[J]. 遥感学报, 2004, 8(2): 128-136.Huang Lisheng, Wang Zhensong, and Zheng Tianyao. A fast algorithm based on FFT used in simulation of SAR return wave signal[J]. Journal of Remote Sensing, 2004, 8(2): 128-136.
[20]	Moving and Stationary Target Acquisition and Recognition (MSTAR) Public Dataset[OL]. https://www.sdms.afrl. af.mil/datasets/mstar/.