针对舰船通信网络流量易受噪声成分影响，导致流量异常检测精度下降问题，提出基于机器学习的舰船通信网络流量异常检测方法。该方法使用基于小波变换的网络流量预处理方法，细化舰船通信网络原始流量数据，由小波阈值将细化后流量数据进行去噪处理后，通过基于机器学习的流量异常检测模型，以前向传播训练、反向传播训练的方式，训练稳定的卷积循环神经网络，将去噪后流量数据样本输入网络中，分类检测通信网络流量数据是否异常。实验结果显示：所提方法有效去除舰船通信网络流量噪声成分后，可提高舰船通信网络流量异常检测精度，无错检情况，且检测范围更全面。
To address the issue of reduced accuracy in detecting traffic anomalies in ship communication networks due to the susceptibility of noise components, a machine learning based method for detecting traffic anomalies in ship communication networks is studied. This method uses a network traffic preprocessing method based on wavelet transform to refine the original traffic data of the ship communication network. After denoising the refined traffic data using wavelet thresholding, a machine learning based traffic anomaly detection model is used to train a stable convolutional recurrent neural network through forward propagation and backward propagation training. The denoised traffic data samples are input into the network, classify and detect whether communication network traffic data is abnormal. The experimental results show that the proposed method can effectively remove the noise component of ship communication network traffic, improve the accuracy of ship communication network traffic anomaly detection, have no false detections, and have a more comprehensive detection range.
2023,45(21): 213-216 收稿日期：2023-4-19
DOI：10.3404/j.issn.1672-7649.2023.21.042
分类号：TN915
基金项目：中央高校基本科研业务费专项资金创新团队资助计划项目（ZY20180125）
作者简介：潘志安（1983-），男，硕士，副教授，研究方向为软件定义网络及计算机网络
参考文献：
[1] 李华峰, 刘炎, 徐凯, 等. 基于异构网络的物联网海洋大气腐蚀加速实验平台[J]. 中国舰船研究, 2021, 16(4): 224-231.LI Hua-feng, LIU Yan, XU Kai, et al. Marine atmospheric accelerated experimental corrosion platform based on IoT technology through heterogeneous network[J]. Chinese Journal of Ship Research, 2021, 16(4): 224-231.
[2] 孟永伟, 秦涛, 赵亮, 等. 利用残差分析的网络异常流量检测方法[J]. 西安交通大学学报, 2020, 54(1): 42-48+84.MENG Yong-wei, QIN Tao, ZHAO Liang, et al. Network Anomaly Detection Method Based on Residual Analysis[J]. Journal of Xi'an Jiaotong University, 2020, 54(1): 42-48+84.
[3] 麻文刚, 张亚东, 郭进. 基于LSTM与改进残差网络优化的异常流量检测方法[J]. 通信学报, 2021, 42(5): 23-40.MA Wen-gang, ZHANG Ya-dong, GUO Jin. Abnormal traffic detection method based on LSTM and improved residual neural network optimization[J]. Journal on Communications, 2021, 42(5): 23-40.
[4] 董书琴, 张斌. 面向不平衡数据的网络流量异常检测方法[J]. 系统仿真学报, 2021, 33(3): 679-689.
[5] 展鹏, 陈琳, 曹鲁慧, 等. 基于特征符号表示的网络异常流量检测算法[J]. 浙江大学学报(工学版), 2020, 54(7): 1281-1288.ZHAN Peng, CHEN Lin, CAO Lu-hui, et al. Network traffic anomaly detection based on feature-based symbolic representation[J]. Journal of Zhejiang University (Engineering Science), 2020, 54(7): 1281-1288.
