大型舰船通信网络入侵检测方法研究

Research on intrusion detection method for large ship communication network

针对传统的入侵检测方法一直存在入侵检测误差大、收敛性差的问题,提出基于惩罚函数和多元回归分析数学模型的大型舰船通信网络入侵检测方法,通过确定入侵节点区域面积计算舰船通信网络入侵节点聚集度,引入一个惩罚函数,再采用决策树分类法对舰船通信网络入侵节点进行分类。采用均方差作为标准测度函数,获取均方差之和,并通过构建多元回归分析数学模型,实现对大型舰船通信网络入侵的检测。实验结果对比可知,采用改进检测方法时,其检测误差及算法的收敛性,一直优于传统检测方法,具有一定的实用性。

According to the traditional intrusion detection methods have intrusion detection error and convergence problem, based on the penalty function and the multiple regression analysis of mathematical model of ship communication network intrusion detection, intrusion node by determining the area calculation of ship communication network intrusion node aggregation, introduce a penalty function, then using decision tree classification method of ship communication network intrusion node. The mean square error is used as the standard measure function to obtain the sum of the mean square error, and the mathematical model of multiple regression analysis is used to realize the detection of the communication network intrusion of large ships. The experimental results show that the improved detection method, the detection error and the convergence of the algorithm, has been better than the traditional detection method, has a certain practicality.