宽带信号匹配滤波的GPU实现及性能优化

doi:10.3969/j.issn.1001-2400.2015.03.023

J4 ›› 2015, Vol. 42 ›› Issue (3): 135-140+191.doi: 10.3969/j.issn.1001-2400.2015.03.023

宽带信号匹配滤波的GPU实现及性能优化

周航;蔡志明;王希敏

(海军工程大学电子工程学院，湖北武汉 430033)

收稿日期:2014-05-04 修回日期:2014-05-28 出版日期:2015-06-20 发布日期:2015-07-27
通讯作者: 周航
作者简介:周航(1985-)，男，海军工程大学博士研究生，E-mail：zhh06@163.com．
基金资助:
国家自然科学基金资助项目(51009146)

Implementation and optimization of the wideband matched filter on the GPU

ZHOU Hang;CAI Zhiming;WANG Ximin

(School of Electronic Engineering, Naval Univ. of Engineering, Wuhan 430033, China)

Received:2014-05-04 Revised:2014-05-28 Online:2015-06-20 Published:2015-07-27
Contact: ZHOU Hang

摘要/Abstract

摘要：

从宽带相关的角度推导了基于小波变换的匹配滤波算法及基于快速傅里叶变换(FFT)算法，并分析了算法复杂度，提出了基于图形处理器(GPU)的可配置宽带匹配滤波的软件实现和理论预测与函数实测结合的优化方法．通过优化线程块的维度、绑定纹理寄存器来改进内核函数性能，再使用计算统一设备架构(CUDA)库来降低FFT与极值搜索的时延，并进行了性能优化设计．在性能测试中，文中方法在GPU平台的实现相比8核CPU平台的实现具有3.3倍加速比，其处理时延能够满足宽带匹配滤波的实时性需求．

关键词: 信号处理, 并行计算, 图形处理器, 程序优化, 连续小波变换

Abstract:

The fine estimation of wideband ambiguity, which has a sharp main ridge, requires large amounts of searching on the time-scale. That desperately needs the well-optimized software on high performance hardware. In terms of wideband correlation, the matched filter based on the CWT and its fast algorithm based on the FFT are studied, and furthermore its complexity is analyzed. Then a reconfigurable implementation on the GPU is proposed, and a method of optimization that combines analysis with testing is proposed. By optimizing the dimension of the thread block and utilizing texture memory, the time of the kernel is reduced; the CUDA library is introduced, so the delays of the FFT and maximum searching are reduced. In comparison with the method in the 8-core CPU, the proposed method improves the overall performance up to 3.3 times. The speed can meet the challenge of real-time processing of the wideband matched filter.

Key words: signal processing, parallel computing, graphics processing unit (GPU), program optimization, continuous wavelet transform (CWT)

周航;蔡志明;王希敏. 宽带信号匹配滤波的GPU实现及性能优化[J]. J4, 2015, 42(3): 135-140+191.

ZHOU Hang;CAI Zhiming;WANG Ximin. Implementation and optimization of the wideband matched filter on the GPU[J]. J4, 2015, 42(3): 135-140+191.

参考文献

［1］秦华, 周沫, 察豪, 等. 软件雷达信号处理的多GPU并行技术［J］. 西安电子科技大学学报, 2013, 40(3): 145-151.
Qin Hua, Zhou Mo, Cha Hao, et al. Research on Multi-GPU Parallel Technology in Software Radar Signal Processing ［J］. Journal of Xidian University, 2013, 40(3): 145-151.
［2］蔡志明, 王希敏. 软件声纳的概念与趋势［J］. 声学技术, 2007, 26(5): 968-971.
Cai Zhiming, Wang Ximin. Concept and Development of the Software-defined Sonar［J］.Technical Acoustics, 2007, 26(5): 968-971.
［3］王军, 李亚安. 基于梅林变换的宽带水声信号处理研究［J］. 兵工学报, 2007, 28(1): 87-90.
Wang Jun, Li Yaan. A Research on Wideband Underwater Acoustic Signal Processing Based on Mellin Transform［J］. Acta Armamentarii, 2007, 28(1): 87-90.
［4］杨定洲. 基于小波变换的水下信号处理研究［D］. 西安: 西北工业大学, 2005.
［5］ Wong T T, Leung C S, Heng P A, et al. Discrete Wavelet Transform on Consumer-level Graphics Hardware［J］. IEEE Transactions on Multimedia, 2007, 9(3): 668-673.
［6］ Suren C, Alessandro M, Andrew H, et al. A GPU-based Architecture for Real-time Data Assessment at Synchrotron Experiments［J］. IEEE Transactions on Nuclear Science, 2011, 58(4): 1447-1455.
［7］ Wu Jing, JaJa J, Balaras E. An Optimized FFT-based Direct Poisson Solver on CUDA GPUs［J］. IEEE Transactions on Parallel and Distributed Systems, 2014, 25(3): 550-559.
［8］ Liu Lifeng, Liu Meilin, Wang Chongjun. An Optimized GP-GPU Warp Scheduling Algorithm for Sparse Matrix-vector Multiplication［C］//IEEE Eighth International Conference on Networking, Architecture and Storage. Washington: IEEE, 2013: 222-231.
［9］ Guo Ping, Wang Liqiang, Chen Po. A Performance Modeling and Optimization Analysis Tool for Sparse Matrix-vector Multiplication on GPUs［J］. IEEE Transactions on Parallel and Distributed Systems, 2014, 25(5): 1112-1123.
［10］李晓敏, 侯朝焕, 鄢社锋. 一种基于GPU的主动声纳宽带信号处理实时系统［J］. 传感技术学报, 2011, 24(9): 1279-1283.
Li Xiaomin, Hou Chaohuan, Yan Shefeng. A Real-time Signal Processing System of Broadband Active Sonar Based on GPU［J］. Chinese Journal of Sensors and Actuators, 2011, 24(9): 1279-1283.
［11］ Garland M, Le Grand S, Nickolls J, et al. Parallel Computing Experiences with CUDA［J］. IEEE Micro, 2008, 28(4): 13-27.

[1]	翟畅,林中朝,赵勋旺,张玉. 一种使用八叉树的半空间MLFMA区域分解算法[J]. 西安电子科技大学学报, 2021, 48(6): 144-150.
[2]	陶毅,丁丽. 利用共轭对称性的数字IQ频域校准方法[J]. 西安电子科技大学学报, 2021, 48(2): 181-189.
[3]	杨杰,蒋俊正. 利用联合图模型的传感器网络数据修复方法[J]. 西安电子科技大学学报, 2020, 47(1): 44-51.
[4]	董刚,黄嵩人,陈迪平,杨翠灵,易峰. 一种内嵌于DSP芯片的高耐压ECAN驱动器[J]. 西安电子科技大学学报, 2019, 46(1): 137-142.
[5]	校松, 陈辉, 倪萌钰, 倪柳柳, 张佳佳. 利用最小冗余对称阵列的近场源定位算法[J]. 西安电子科技大学学报, 2018, 45(6): 123-129.
[6]	宫健;楼顺天;张伟涛. 一种强干扰条件下阵列天线波达方向估计方法[J]. 西安电子科技大学学报, 2018, 45(1): 168-172.
[7]	廖瑞乾;许京伟;廖桂生. 近场散射环境下的稳健自适应波束形成[J]. 西安电子科技大学学报, 2017, 44(5): 184-188.
[8]	姚亚峰;冯中秀;陈朝. 超低时延免迭代CORDIC算法[J]. 西安电子科技大学学报, 2017, 44(4): 162-166.
[9]	任计林;车书玲;郑征. 规则LDPC码在GPU上的加速译码[J]. 西安电子科技大学学报, 2017, 44(3): 25-30.
[10]	夏璞;刘学斌;闫鹏. 偏振透雾成像系统的设计与实验分析[J]. 西安电子科技大学学报, 2016, 43(2): 95-101.
[11]	刘镇弢;李涛;黄虎才;韩俊刚;沈绪榜 . 一种用于实时图像处理的众核结构设计[J]. J4, 2015, 42(2): 95-101.
[12]	白冰;牛中奇. 时域有限差分法中的GPU加速高效CPML方案[J]. J4, 2015, 42(1): 194-199+212.
[13]	田玉玲. 面向异常检测的时间序列树突状细胞算法[J]. J4, 2014, 41(4): 144-150.
[14]	赵宝玮;相里斌;吕群波;张桂峰;曾晓茹;郭文记. FPGA和多DSP系统的并行RX探测算法[J]. J4, 2014, 41(3): 152-156.
[15]	王凌;刘坚强;李国林;张真;杨继坤. 用旋转不变技术实现互耦和通道不一致的联合抑制[J]. J4, 2014, 41(1): 182-188.

宽带信号匹配滤波的GPU实现及性能优化

Implementation and optimization of the wideband matched filter on the GPU

PDF (PC)

赞

可视化

被引次数

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

Metrics

本文评价

推荐阅读 10