摘要
以提升光纤网络威胁感知准确性为目标,提出基于多信息融合的光纤网络威胁智能感知方法。该架构由信息获取层与量化感知层共同组成,信息获取层:利用多个传感器提取光纤网络多路信号数据,数据经去噪后,利用人工神经网络对降噪后的数据实施多信息融合;量化感知层:提取融合信息中网络安全影响度最高的数据,通过量化感知技术将信息转化为威胁势态,利用抽象能力创建势态的变化过程后,经过持续推理,展现完整的网络威胁势态,实现网络威胁智能感知。实验结果表明:该方法的数据融合计算检测率和融合率较高,误警率较低,数据融合计算能力强;可有效感知不同服务器、不同主机以及网络的威胁势态,且感知正确率始终高于97%。
In order to improve the accuracy of optical fiber network threat perception,an intelligent threat perception method based on multi-information fusion is proposed. The architecture consists of an information acquisition layer and a quantitative perception layer. In the information acquisition layer,multi-channel signal data of an optical fiber network are extracted by using a plurality of sensors,and after the data is denoised,multi-information fusion is carried out on the denoised data by using an artificial neural network;Quantitative perception layer: Extract the data with the highest degree of network security influence from the fusion information,transform the information into threat situation through quantitative perception technology,create the change process of situation by using abstract ability,and show the complete network threat situation through continuous reasoning,so as to realize intelligent perception of network threat. The experimental results show that this method has high detection rate and fusion rate,low false alarm rate and strong data fusion calculation ability. It can effectively sense the threat situation of different servers,different hosts and networks,and the sensing accuracy is always higher than 97%.
作者
李秀峰
王崇霞
LI Xiufeng;WANG Chongxia(Department of Computing,Changzhi University,Changzhi Shanxi 046011,China)
出处
《激光杂志》
CAS
北大核心
2021年第10期138-142,共5页
Laser Journal
基金
山西省高等学校教学改革创新项目(No.J2020325)。
关键词
信息融合
光纤
威胁
智能
感知
人工网络
information fusion
optical fiber
threat
intelligence
perception
artificial network