基于CUDA架构的混合测向算法研究

金莉; 宋万杰; 施治国

引用本文:	金莉，宋万杰，施治国. 基于CUDA架构的混合测向算法研究[J]. 雷达科学与技术, 2018, 16(2): 133-138.[点击复制]
	JIN Li, SONG Wanjie, SHI Zhiguo. Study on Hybrid Direction Finding Algorithm Based on CUDA Framework[J]. Radar Science and Technology, 2018, 16(2): 133-138.[点击复制]

【打印本页】【下载PDF全文】【查看/发表评论】【下载PDF阅读器】【关闭】

←前一篇|后一篇→

过刊浏览高级检索

本文已被：浏览 6105次下载 979次	码上扫一扫！
分享到：微信更多字体:加大+\|默认\|缩小-
基于CUDA架构的混合测向算法研究
金莉，宋万杰，施治国
1.西安电子科技大学雷达信号处理国家重点实验室，陕西西安710071;2.航天南湖电子信息技术股份有限公司，湖北荆州434000

摘要:

采用数字波束形成算法与合成导向矢量算法相结合的测高算法，提高了测角精度和测角速率。另外，为了进一步满足系统实时性要求，提出采用基于计算统一设备架构为平台完成雷达测角的方案，该方案利用并行计算平台和异构编程模型，首先在通用处理计算机上用CPU完成预处理，并控制信号处理的任务调度和负载分配，图形处理器实现数字波束形成算法和合成导向矢量算法，并与CPU进行比较。实验结果表明，利用计算统一设备架构技术实现的测高算法取得了比CPU算法更高的运算效率。

关键词: 数字波束形成算法合成导向矢量算法并行异构编程模型计算统一设备架构

DOI：10.3969/j.issn.1672-2337.2018.02.003

分类号:TN958；TN957.51

基金项目:国家自然科学基金（No.61571349）

Study on Hybrid Direction Finding Algorithm Based on CUDA Framework

JIN Li, SONG Wanjie, SHI Zhiguo

1.National Key Lab of Radar Signal Processing, Xidian University, Xi’an 710071, China;2.Aerospace Nanhu Electronic Information Technology Company Limited, Jingzhou 434000, China

Abstract:

In this paper, a direction finding algorithm combining DBF algorithm with SVML algorithm is proposed, which can improve the precision of angle measurement and angular velocity. In order to meet the real-time requirement of the system, we present a radar angle measurement scheme based on CPU+GPU heterogeneous parallel architecture. In other words, on the general processing computer, CPU is used to complete the preprocessing and control the task scheduling and load allocation of signal processing, and GPU is used to realize DBF algorithm and SVML algorithm. Finally, a comparison with CPU is made. The experimental results show that the algorithm achieved by CUDA has higher efficiency than CPU.

Key words: DBF algorithm SVML algorithm parallel CPU+GPU compute unified device architecture (CUDA)