系统工程与电子技术 ›› 2026, Vol. 48 ›› Issue (6): 1800-1808.doi: 10.12305/j.issn.1001-506X.2026.06.02

• 电子技术 • 上一篇    下一篇

基于改进遗传算法的非同步测量低频声源定位

韦娟1,*(), 陈茂楠1(), 冯鹏1(), 宁方立2()   

  1. 1. 西安电子科技大学通信工程学院,陕西 西安 710071
    2. 西北工业大学机电学院,陕西 西安 710072
  • 收稿日期:2025-07-03 修回日期:2025-08-12 出版日期:2026-06-25 发布日期:2025-11-25
  • 通讯作者: 韦娟 E-mail:weijuan@xidian.edu.cn;995455137@qq.com;1638267477@qq.com;ningfl@nwpu.edu.cn
  • 作者简介:陈茂楠(2001—),女,硕士研究生,主要研究方向为声源定位
    冯 鹏(1999—),男,硕士研究生,主要研究方向为信息处理、声源定位
    宁方立(1974—),男,教授,博士,主要研究方向为声源定位
  • 基金资助:
    国家自然科学基金(52475132);陕西省重点研发计划(2024GX-ZDCYL-01-16);航空科学基金(20200015053001);西安市重点产业链技术攻关(23ZDCYJSGG0006-2023)资助课题

Low-frequency sound source localization based on improved genetic algorithm with non-synchronous measurement

Juan WEI1,*(), Maonan CHEN1(), Peng FENG1(), Fangli NING2()   

  1. 1. School of Telecommunications Engineering,Xidian University,Xi’an 710071,China
    2. School of Mechanical Engineering,Northwestern Polytechnical University,Xi’an 710072,China
  • Received:2025-07-03 Revised:2025-08-12 Online:2026-06-25 Published:2025-11-25
  • Contact: Juan WEI E-mail:weijuan@xidian.edu.cn;995455137@qq.com;1638267477@qq.com;ningfl@nwpu.edu.cn

摘要:

针对传统声源定位方法在低频条件下定位效果差的问题,提出一种基于遗传算法的非同步测量声源定位算法。该算法根据声源平面划分的网格点数确定染色体长度,采用二进制将解空间编码成有限长度的字符串,并根据每个个体在非同步测量下的缺省互谱矩阵定义适应度函数,进行优化搜索。为提高定位效率,结合常规波束形成方法对搜索空间进行预处理,从而缩小搜索范围。实验结果表明,所提算法在低频、低信噪比条件下仍能实现高精度定位,并在多声源环境中展现出良好的鲁棒性与抗噪性能。

关键词: 遗传算法, 非同步测量, 声源定位, 波束形成

Abstract:

To address the issue of poor localization performance of traditional sound source localization methods in low-frequency conditions, a non-synchronous measurement-based sound source localization algorithm using genetic algorithm is proposed. This algorithm determines the chromosome length based on the number of grid points dividing the sound source plane, encodes the solution space into finite-length strings using binary representation, and defines the fitness function according to the default cross-spectral matrix of each individual in non-synchronous measurements, thereby enabling optimized search. To improve localization efficiency, the conventional beamforming method is incorporated to preprocess the search space for narrowing the search range. Experimental results demonstrate that the proposed algorithm achieves high-precision localization even with low-frequency and low signal-to-noise ratio conditions, while exhibiting strong robustness and noise resistance in multi-source environment.

Key words: genetic algorithm, non-synchronous measurement, sound source localization, beamforming

中图分类号: