为了提高对舰船通信网络异常识别能力，提出基于数据驱动的舰船通信网络异常行为检测方法。采用最短路径和最大覆盖范围寻优方法构建舰船通信网络的节点覆盖模型，通过最小间隔均衡技术对舰船通信网络的信道均衡控制，提取舰船通信网络的信道传输信息特征。对舰船通信网络的行为特征参数分析，结合谱分量融合和融合聚类处理方法，实现对舰船网络异常行为的数据驱动控制。根据数据驱动的图模型参数识别和异常谱特征聚类分析，实现对舰船通信网络异常行为检测。测试结果表明，该方法能够进行舰船通信网络异常行为检测处理，提高信道均衡性能
In order to improve the ability to identify abnormal behaviors in ship communication networks, a data-driven method for detecting abnormal behaviors in ship communication networks is proposed. The node coverage model of the ship communication network is constructed using the shortest path and maximum coverage optimization method. The minimum interval equalization technology is used to control the channel balance of the ship communication network, extract the channel transmission information characteristics of the ship communication network, analyze the behavioral characteristic parameters of the ship communication network, and combine spectral component fusion and fusion clustering processing methods to achieve data-driven control of abnormal behavior in the ship network, Based on data-driven graph model parameter identification and anomaly spectrum feature clustering analysis, abnormal behavior detection of ship communication networks is achieved. Testing has shown that this method detects and processes abnormal behavior in ship communication networks, improving channel equalization performance
2023,45(10): 131-134 收稿日期：2022-10-30
DOI：10.3404/j.issn.1672-7649.2023.10.025
分类号：TN911
基金项目：教育部产学合作协同育人项目项目（220500383190724）
作者简介：莫凡(1989-),男,硕士,讲师,研究方向为计算机网络、移动应用开发及数据挖掘
