Page 152 - 201903

第 38 卷 第 3 期 Vol. 38, No. 3
2019 年 5 月 Journal of Applied Acoustics May, 2019

⋄ 研究报告 ⋄

低信噪比条件下多节点声呐目标跟踪算法 ∗

李 嶷 † 陈新华 郑恩明 方 华 王麟煜

(中国科学院声学研究所 北京 100190)

摘要 低信噪比条件下，单个主动或单个被动声呐节点难以实现目标跟踪。多节点声呐系统希望通过增加探
测节点提高系统探测能力，但其性能提高与否取决于选择合适的数据融合算法。该文利用双节点声呐实验数
据研究低信噪比条件下的目标跟踪方法，采用 “结合置信度水平的表决融合” 算法对多个探测节点得到的数据
进行融合，既考虑了目标回波信号的信噪比特性，又考虑了目标运动的连续性特征，还考虑了各节点探测结果
的决策优化，最终实现较高精度的目标跟踪。算法实现了数据级、特征级和决策级的统一、融合，通过对判决依
据进行量化、分层，简化判决的复杂性。实验数据处理结果表明，该方法能较好地解决低信噪比条件下多节点
声呐目标跟踪问题，目标跟踪精度较高。
关键词 信号处理，多节点声呐系统，数据融合，目标跟踪
中图法分类号: TB56 文献标识码: A 文章编号: 1000-310X(2019)03-0434-06
DOI: 10.11684/j.issn.1000-310X.2019.03.021

Target tracking algorithm for multistatic sonar with low signal noise ratio

LI Yi CHEN Xinhua ZHENG Enming FANG Hua WANG Linyu

(Institute of Acoustics, Chinese Academy of Sciences, Beijing 100190, China)

Abstract It’s difficult for single active or single passive sonar to track underwater targets with low signal
noise ratio (SNR). So adding more detecting nodes into multistatic sonar system is expected. But whether
it can improve the detection capabilities of the sonar system lies on designing appropriate fusion algorithm.
Experimental data of bistatic sonar system under low SNR was used to investigate target tracking problems.
And a voting fusion algorithm combined with confidence level was put forward. This algorithm takes into
account not only the SNR of target echoes, but also the continuity characteristic of target motion, and also
the decision methods of the detection results from many nodes. Finally, higher precision tracking results was
achieved. This algorithm realizes the unification and fusion of data level, feature level and decision level.
By quantifying the decision criterion, the judgments of target tracks were simplified. Processing results of
experimental data show that the proposed method can solve the target tracking problem detected by multistatic
sonar with lower SNR, and the target tracking accuracy is higher.
Key words Signal processing, Multistatic sonar system, Data fusion, Target tracking

2018-10-09 收稿; 2018-12-31 定稿
∗ 国防科技创新特区项目
作者简介: 李嶷 (1973- ), 女, 四川人, 博士, 研究方向: 水声信号处理。
† 通讯作者 E-mail: liyi_731973@aliyun.com