基于超融合架构的恶意网络行为检测仿真  

Simulation of Malicious Network Behavior Detection Based on Hyper-Converged Architecture

作　　者：祝丹[1] 卢凌燕[2] ZHU Dan;LU Ling-yan(College of Mathematics and Science,Wuhan Institute of Technology,Wuhan Hubei 430205,China;School of Computer Science and Engineering,Wuhan Institute of Technology,Wuhan Hubei 430205,China)

机构地区：[1]武汉工程大学数理学院,湖北武汉430205 [2]武汉工程大学计算机科学与工程学院,湖北武汉430205

出　　处：《计算机仿真》2023年第10期404-407,510,共5页Computer Simulation

摘　　要：恶意网络行为形式多样,攻击者不断创新和改进手段,加大了检测难度,为了提升恶意网络行为检测结果准确性,提升防御能力,提出一种基于超融合架构的恶意网络行为检测方法。对数据进行离散化处理,确定网络中模糊数据的关联规则。采用基于模糊关联规则的数据挖掘方法获取经过整理的网络数据,同时通过高可用模型提供的信息,对超融合基础架构进行优化,采用优化后的架构对网络恶意行为特征进行提取,对网络流量特征和信息熵特征进行加权处理,构建恶意网络行为检测模型,通过模型最终实现恶意网络行为检测。仿真结果表明,所提方法可以获取高精度的检测结果,具有较强的防御能力。There are various forms of malicious network behavior,and attackers constantly innovate and improve their methods,increasing the difficulty of detection.In order to improve the accuracy of malicious network behavior detection results and enhance defense capabilities,a malicious network behavior detection method based on hyperfusion architecture is proposed.The data was discretized at first in order to determine the association rules of fuzzy data in network.Then,a data mining method based on fuzzy association rules was adopted to obtain the sorted network data.In the meanwhile,the hyper-converged architecture was optimized through the information provided by the high availability model.After that,the optimized architecture was used to extract the characteristics of network malicious behavior.In addition,the characteristics of network traffic and information entropy were weighted.Finally,a model was built for detecting malicious network behaviors.Simulation results prove that the proposed method can obtain high-precision detection results,with strong defense ability.

关 键 词：超融合架构 恶意网络行为 关联规则 数据挖掘 高可用模型 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]