传统频域滤波与时域分析方法难以有效处理非平稳干扰的时变特性与多源耦合问题。分析船舶局域网络干扰的时域、频域特征，构建基于小波分析的“信号预处理—特征提取—源定位”一体化框架，提出自适应小波阈值去噪算法，提取能量熵、频率重心、时域突变点密度构成多维度特征向量，实现干扰类型的精准分类。结合小波时频局部化特性与TDOA技术，构建干扰源时空定位模型，仿真试验表明，该方法对高斯噪声、脉冲干扰和混合干扰的信噪比提升分别达15.8、18.3、16.2 dB，验证了本文方法的有效性与工程应用价值。
Traditional frequency-domain filtering and time-domain analysis methods are difficult to effectively handle the time-varying characteristics and multi-source coupling problems of non-stationary interference. Analyze the time-domain and frequency-domain characteristics of the interference in the ship's local area network, construct an integrated framework of "signal preprocessing - feature extraction - source location" based on wavelet analysis, propose an adaptive wavelet threshold denoising algorithm, extract energy entropy, frequency center of gravity, and density of time-domain mutation points to form multi-dimensional feature vectors, and achieve precise classification of interference types. By combining the wavelet time-frequency localization characteristics with TDOA technology, a spatio-temporal location model of the interference source was constructed. Simulation experiments show that the signal-to-noise ratio improvement of this method for Gaussian noise, pulse interference and mixed interference reaches 15.8, 18.3, 16.2 dB respectively, verifying its effectiveness and engineering application value in the mining of interference information in ship networks.
2025,47(16): 173-176 收稿日期：2025-3-20
DOI：10.3404/j.issn.1672-7649.2025.16.027
分类号：U667.65
基金项目：山西省教育科学“十四五”规划课题（GH-230252）
作者简介：李慧姝(1981-),女,硕士,副教授,研究方向为人工智能及信息化
参考文献：
[1] 吴大鹏, 赵莹, 熊余, 等. 基于小波神经网络的告警信息相关性挖掘策略[J]. 电子与信息学报, 2014, 36(10): 2379-2384.
WU D P, ZHAO Y, XIONG Y, et al. Correlation mining strategy of alarm information based on wavelet neural network[J]. Journal of Electronics and Information Technology, 2014, 36(10): 2379-2384.
[2] 周慧, 魏霖静, 李玥. 改进小波变换的船舶射频信号降噪和识别[J]. 舰船科学技术, 2023, 45(6): 162-165.
ZHOU H, WEI L J, LI Y. Noise reduction and recognition of ship radio frequency signals based on improved wavelet transform[J]. Ship Science and Technology, 2023, 45(6): 162-165.
[3] 陈玲萍, 杨呈永. 面向高维异常数据挖掘的小波变换算法优化[J]. 计算机仿真, 2025, 42(1): 462-465+472.
CHEN L P, YANG C Y. Optimization of wavelet transform algorithm for high-dimensional anomaly data mining[J]. Computer Simulation, 2025, 42(1): 462-465+472.
[4] 朱哲华. 基于小波神经网络的船舶电气故障诊断[J]. 舰船科学技术, 2023, 45(20): 172-175.
ZHU Z H. Fault diagnosis of marine electrical equipment based on wavelet neural network[J]. Ship Science and Technology, 2023, 45(20): 172-175.
[5] 窦新宇, 贾兆旻. 基于小波分析技术的船舶测试信号分析[J]. 舰船科学技术, 2020, 42(10): 181-183.
DOU X Y, JIA Z M. Analysis of ship test signals based on wavelet analysis technology[J]. Ship Science and Technology, 2020, 42(10): 181-183.
[6] 赵雨佳, 邢传玺, 姜佳圆, 等. 利用船舶噪声线谱的TDOA和FDOA定位方法研究[J]. 云南民族大学学报(自然科学版), 2024, 33(5): 638-644.
ZHAO Y J, XING C X, JIANG J Y, et al. Research on TDOA and FDOA location methods using ship noise line spectrum[J]. Journal of Yunnan Minzu University (Natural Sciences Edition), 2024, 33(5): 638-644.
