西安电子科技大学学报 ›› 2024, Vol. 51 ›› Issue (6): 1-9.doi: 10.19665/j.issn1001-2400.20241003

• 信息与通信工程 •    下一篇

一种非合作多目标的分布式定位算法

李文刚1(), 王乾雄2(), 黄郡3(), 慈国辉4(), 翟肖童2()   

  1. 1.西安电子科技大学 通信工程学院,陕西 西安 710071
    2.西安电子科技大学 广州研究院,广东 广州 510555
    3.国防科技大学 电子对抗学院,安徽 合肥 230031
    4.中国电子科技集团公司第五十四研究所,河北 石家庄 050091

Non-cooperative multi-target distributed localization algorithm

LI Wengang1(), WANG Qianxiong2(), HUANG Jun3(), CI Guohui4(), ZHAI Xiaotong2()   

  1. 1. School of Telecommunications Engineering,Xidian University,Xi’an 710071,China
    2. School of Guangzhou Institude of Technology,Xidian University,Guangzhou 510555,China
    3. College of Electronic Engineering,National University of Defense Technology,Hefei 230031,China
    4. The 54th Research Institute of CETC,Shijiazhuang 050081,China
  • Received:2023-12-07 Online:2024-10-23 Published:2024-10-23

摘要:

随着移动通讯领域的蓬勃发展,现代战争中对于远距离目标的探测更加注重隐蔽性和安全性。在这一背景下,采用被动接收信号的方式进行非合作多目标定位的研究受到广泛关注,然而以被动的方式对多个非合作目标进行定位时,信号数据之间的匹配关系极大程度上影响了各目标位置估计结果的精确度。针对数据关联关系的复杂性和多样性的问题,提出了一种非合作多目标的分布式定位算法。算法通过获取信号计算角度和强度量测,并构建量测数据对,筛选候选参考点以降低量测数据关联复杂度,以量测数据对的匹配概率分析构建代价函数,通过选取最小代价点所量测信息作为该目标的最匹配量测,并通过聚类分析构建最终代理点集合解算各待测目标位置。仿真结果表明,在相同场景中所提算法能将目标匹配成功率提高约26.3%,运算速度提高约16.8%,且在6 km×4 km的区域中将平均定位误差控制在约17.41 m,能够快速、准确地进行多目标位置估计。

关键词: 非合作多目标, 分布式定位, 数据关联, 代价函数, 聚类分析

Abstract:

With the vigorous development of mobile communication,the detection of long-distance targets in modern warfare pays more attention to concealment and security.In this context,the research on non-cooperative multi-target positioning by passive receiving signals has attracted wide attention.However,passive localization of multiple non-cooperative targets is greatly affected by the matching relationship between signal data,which influences the accuracy of target position estimation.To address the complexity and diversity of data matching relationships,this paper proposes a distributed data matching localization algorithm for non-cooperative multi-targets.The algorithm obtains angle and intensity measurements from signal calculations,constructs measurement data pairs,and selects candidate reference points to reduce measurement data matching complexity.It then constructs a cost function through the probability analysis of measurement data pair matching,selects the measurement information of the minimum cost point as the best match for the target,and calculates the position of each target by the clustering analysis of final proxy point sets.Simulation results show that the proposed algorithm can increase the target matching success rate by over 26.3% and the computational speed by 16.8% in the same scenario,and can achieve fast and accurate multi-target localization with an average positioning error of about 17.41 m in a region of 6 km×4 km.

Key words: non-cooperative multi-target, distributed localization, data association, cost functions, cluster analysis

中图分类号: 

  • TN96