J4 ›› 2011, Vol. 38 ›› Issue (2): 180-186.doi: 10.3969/j.issn.1001-2400.2011.02.032

• 研究论文 • 上一篇    下一篇

拓扑序列航迹相关的高效修正算法

  吴泽民1;蒋叶金2;任姝婕3
  

  1. (1. 解放军理工大学 通信工程学院,江苏 南京  210007;
    2. 中国人民解放军73683部队,福建 福州  350102;
    3. 解放军理工大学 理学院,江苏 南京  210007)
  • 收稿日期:2010-03-28 出版日期:2011-04-20 发布日期:2011-05-26
  • 通讯作者: 吴泽民
  • 作者简介:吴泽民(1973-),男,副教授,博士,E-mail: wuzemin_ice@163.com.
  • 基金资助:

    国家863计划高技术资助项目(2008AA01Z216);国家973资助项目(2009CB3020402)

Effectively modified topology sequence track correlation algorithm

WU Zemin1;JIANG Yejin2;REN Shujie3   

  1. (1. Inst. of Communication, PLA Univ. of Science and Tech., Nanjing  210007, China;
    2. Unit 73683 of PLA, Fuzhou  350102, China;
    3. Inst. of Sciences, PLA Univ. of Science and Tech., Nanjing  211101, China)
  • Received:2010-03-28 Online:2011-04-20 Published:2011-05-26
  • Contact: WU Zemin

摘要:

针对拓扑序列航迹相关法为适应较大系统误差而计算量过大的问题,首先证明了存在系统误差时,不同传感器的拓扑序列满足近似线性变换关系,然后引入空间点集的奇异值分解(SVD)匹配算法对拓扑序列进行直接计算,通过检验得到的线性变换参数来判定航迹的相关性.SVD算法是一种高效的航迹相关算法,尤其适用于远距离目标的拓扑序列近似.通过仿真,验证了SVD算法的有效性,不但计算时间降低了90%以上,而且避免了拓扑序列法在进行角度和径向距离步进匹配时步长选择的难题,提高了航迹的相关成功率.

关键词: 数据融合, 航迹相关, 奇异值分解, 拓扑

Abstract:

To release the huge calculation load for the track correlation algorithm based on the topology sequence facing large sensor system errors, the approximate linear transformation relationship between topology sequences from different sensors is firstly discussed, where sensor system errors are considered. The SVD method for space points set matching problem is directly used to estimate the linear transformation. A decision for topology sequences matching is made after validating the linear parameters. The SVD based track correlation algorithm is efficient, especially for far-away targets. The efficiency of the SVD based algorithm is validated by simulation. Because of removing the dilemma of choosing the angle and range search step sizes for the basic topology sequence algorithm, not only is the computational time reduced about 90%, but also the successful rate of correlation is improved.

Key words: data fusion, track correlation, SVD, topology