J4 ›› 2015, Vol. 42 ›› Issue (1): 104-111.doi: 10.3969/j.issn.1001-2400.2015.01.017

• 研究论文 • 上一篇    下一篇

CUDA架构下外辐射源雷达杂波抑制并行算法

武勇;王俊;张培川;曹运合
  

  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)
  • 收稿日期:2013-10-17 出版日期:2015-02-20 发布日期:2015-04-14
  • 通讯作者: 武勇
  • 作者简介:武勇(1987-),男,西安电子科技大学博士研究生,E-mail: wuyongxidian@163.com.
  • 基金资助:

    国家自然科学基金资助项目(61372136)

Parallel clutter suppression algorithm for passive radar in CUDA

WU Yong;WANG Jun;ZHANG Peichuan;CAO Yunhe   

  1. (National Key Lab. of Radar Signal Processing, Xidian Univ., Xi'an  710071, China)
  • Received:2013-10-17 Online:2015-02-20 Published:2015-04-14
  • Contact: WU Yong

摘要:

为了解决用于杂波抑制的直接矩阵求逆算法在外辐射源雷达工程应用中实时处理困难的问题,提出了分段并行的直接矩阵求逆算法.充分考虑了直接矩阵求逆算法的特点,将整个运算块分解为多个规模相同的子运算块,在计算统一设备架构下对这些子运算块进行并行处理,达到了提高算法处理效率的目的.实验结果表明,在采样率为200kHz,杂波对消阶数为128的条件下,与传统的串行处理算法相比,该算法的运行效率提高了25倍,并成功用于某外辐射源雷达实时信号处理中.

关键词: 外辐射源雷达, 杂波抑制, 直接矩阵求逆算法, 计算统一设备架构, 并行

Abstract:

In order to solve the problem that the direct matrix inverse(DMI) algorithm for clutter suppression can not be real-time processing in the passive radar engineering application, a segmented parallel DMI algorithm(called P-DMI) is proposed. By considering the feature of the DMI algorithm, the entire operation block is broken up into several identical sub-blocks, and in the Compute Unified Device Architecture, these sub-blocks are processed in parallel to achieve the purpose of improving the processing efficiency. Experimental results indicate that this algorithm's computation efficiency is raised more than 25 times compared with the traditional serial algorithm when the sampling rate is 200kHz and the clutter cancelation order is equal to 128. Also, the method has been used in real-time signal processing for passive radar successfully.

Key words: passive radar, clutter suppression, direct matrix inverse, compute unified device architecture, parallel

中图分类号: 

  • TN958
Baidu
map