摘 要:当前检测网络中存在的异常流量是防止异常流量攻击网络的有效策略之一。本文首先建立了网络流量稳态模型,挖掘并剔除了了网络流量中存在的坏数据。然后通过 S 变换及其逆变换重构网络流量数据,提高了检测精度。最后,提取网络流量特征,在此基础上完成了网络流量异常检测。实验结果表明,所提方法可适用于不同类型网络流量的异常检测,具有良好的检测性能。
关键词:大数据挖掘;稳态模型;S 变换;特征提取;网络流量异常检测
DOI:10.19850/j.cnki.2096-4706.2022.24.021
中图分类号:TP393 文献标识码:A 文章编号:2096-4706(2022)24-0082-04
Network Traffic Abnormal Detection Algorithm Based on Big Data Mining
HU Xiaohong
(College of Artificial Intelligence,Wuxi Vocational College of Science and Technology, Wuxi 214028, China)
Abstract: Currently, detecting abnormal traffic existing in the network is one of the effective strategies to prevent abnormal traffic from attacking the network. In this paper, the steady state model of network traffic is established firstly, and the bad data existing in the network traffic is mined and eliminated. Then the network traffic data is reconstructed through S transform and its inverse transform, which improves the detection accuracy. Finally, the network traffic characteristics are extracted, and the network traffic abnormal detection is completed on this basis. The experimental results show that the proposed method can be applied to abnormal detection of different types of network traffic, and has good detection performance.
Keywords: big data mining; steady state model; S transformation; feature extraction; network traffic abnormal detection
参考文献:
[1] 董书琴,张斌 . 基于深度特征学习的网络流量异常检测方法 [J]. 电子与信息学报,2020,42(3):695-703.
[2] 孟永伟,秦涛,赵亮,等 . 利用残差分析的网络异常流量检测方法 [J]. 西安交通大学学报,2020,54(1):42-48+84.
[3] 孙旭日, 刘明峰, 程 辉, 等 . 结合二次特征提取和LSTM-Autoencoder 的网络流量异常检测方法 [J]. 北京交通大学学报,2020,44(2):17-26.
[4] 卢发兴,姚鸿鹤,史浩然 . 基于方差分析变量约减的指令制导回路误差分配方法 [J]. 系统工程与电子技术,2020,42(5):1131-1138.
[5] 赵宗爽,史治宇,张杰 . 基于二次调频小波变换的时变系统物理参数识别 [J]. 机械制造与自动化,2021,50(6):49-51.
[6] 杨挺,张璐,张亚健,等 . 基于信息熵计算模型的电力信息物理系统融合控制方法 [J]. 电力系统自动化,2021,45(12):65-74.
[7] 麻文刚,张亚东,郭进 . 基于 LSTM 与改进残差网络优化的异常流量检测方法 [J]. 通信学报,2021,42(5):23-40.
[8] 周伯阳,郭志民,王延松,等 . 基于多尺度低秩模型的电力无线接入网异常流量检测方法 [J]. 电子学报,2020,48(8):1552-1557.
作者简介:胡晓红(1984.03—),女,汉族,江苏无锡人,实验师,硕士,研究方向:计算机应用。