基于K-means聚类的舰船通信网络异常数据检测  

Detection of abnormal data in ship communication network based on K-means clustering

作　　者：徐胤博 于洋[1] XU Yin-bo;YU Yang(College of Computer and Information Engineering,Tianjin Normal University,Tianjin 300100,China)

机构地区：[1]天津师范大学计算机与信息工程学院,天津300100

出　　处：《舰船科学技术》2023年第16期169-172,共4页Ship Science and Technology

摘　　要：为了解决海上通信环境中的干扰和传输问题,提升舰船通信网络通信质量和可靠性,提出基于K-means聚类的舰船通信网络异常数据检测方法。构建舰船通信网络通信多径信道模型,利用该模型获取舰船通信网络数据。使用基于超窄带滤波的舰船通信网络数据滤波处理方法去除舰船通信网络数据内的干扰噪声,将无噪声的舰船通信网络数据作为输入,使用K-means聚类算法输出舰船通信网络异常数据检测结果。结果表明,该方法采集舰船通信网络数据较为准确,并可有效去除数据内含有的干扰噪声,降低舰船通信网络数据幅值区间,同时可用聚类方式准确检测舰船通信网络异常数据,应用效果较为显著。In order to solve the interference and transmission problems in the maritime communication environment,improve the communication quality and reliability of ship communication networks,a K-means clustering based abnormal data detection method for ship communication networks is proposed.Construct a multipath channel model for ship communication network communication,and use this model to obtain ship communication network data.Using a ship communication network data filtering processing method based on ultra narrow band filtering to remove interference noise within the ship communication network data,the noise free ship communication network data is used as input,and the K-means clustering algorithm is used to output the abnormal data detection results of the ship communication network.The experimental results show that this method is more accurate in collecting ship communication network data,and can effectively remove interference noise contained in the data,reduce the amplitude range of ship communication network data,and accurately detect abnormal data of ship communication network using clustering method,the application effect is significant.

关 键 词：K-MEANS聚类 舰船通信网络 异常数据检测 马氏距离 超窄带滤波 

分 类 号：TN915[电子电信—通信与信息系统]