基于云计算技术的舰船通信网络异常入侵检测研究  

Research on anomaly intrusion detection of ship communication network based on cloud computing technology

作　　者：于隆 尹娜 YU Long;YIN Na(Applied Technology College,Dalian Ocean University,Dalian 116300,China)

机构地区：[1]大连海洋大学应用技术学院,辽宁大连116300

出　　处：《舰船科学技术》2025年第4期173-177,共5页Ship Science and Technology

基　　金：辽宁省教育科学十三五规划课题(JG16EB043);辽宁省教育科学十四五规划课题(JG21BB039);中国成人教育协会“十四五”成人继续教育科研规划课题(2023-621Y)。

摘　　要：舰船行进中通信节点动态变化,网络拓扑结构多变,易受各种入侵,其中部分隐态入侵因隐蔽性强常被误判,导致入侵检测准确度下降。因此,提出基于云计算技术的舰船通信网络异常入侵检测方法。选取Spark平台,借助Apache Spark框架的Map Reduce处理方式,通过主节点对从节点进行异常入侵检测任务的调度、分发。从节点并行运行长短时记忆网络,利用遗忘门与输入门处理信息,更新记忆单元,输出异常入侵检测结果。主节点利用非线性最小二乘法构建全局异常入侵检测模型,综合从节点检测结果,最终输出舰船通信网络异常入侵检测结果。实验结果表明,该方法能够有效检测模糊攻击、泛攻击等多种舰船通信网络异常入侵行为,且网络吞吐量高于3 Mb/s,具有实际应用性。During the movement of ships,communication nodes dynamically change and network topology structures are diverse,making them vulnerable to various intrusions.Some hidden intrusions are often misjudged due to their strong concealment,resulting in a decrease in intrusion detection accuracy.Therefore,a ship communication network anomaly in-trusion detection method based on cloud computing technology is proposed.Select the Spark platform and utilize the Map Reduce processing method of the Apache Spark framework to schedule and distribute anomaly intrusion detection tasks from the master node to the slave nodes.Run long short-term memory networks in parallel from nodes,use forget gates and input gates to process information,update memory units,and output abnormal intrusion detection results.The master node uses nonlinear least squares method to construct a global anomaly intrusion detection model,integrates the detection results of the slave nodes,and finally outputs the anomaly intrusion detection results of the ship communication network.The experiment-al results show that this method can effectively detect various abnormal intrusion behaviors in ship communication networks such as fuzzy attacks and pan attacks,and the network throughput is higher than 3Mbps,which has practical applicability.

关 键 词：云计算技术 舰船通信网络 异常入侵检测 Spark平台 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]