对等覆盖网络传输层异常流量模糊识别仿真
|
摘要
传统的异常流量识别方法存在识别效率较低、误报率较高等问题,提出基于改进极端随机树的对等覆盖网络传输层异常流量模糊识别方法。分别计算不同特征的信息增益率,得到维度较低的特征集。引用随机训练方法对分类模型进行训练,获取应用于重采样数据分类的最优分类器进行网络传输层流量分类。根据分类结果构建异常流量的统计信号模型,将信号处理方法与高阶量检测算法两者相结合,利用幅频响应特征进行配准,实现对等覆盖网络传输层异常流量识别。实验结果表明,所提方法能够有效提高识别效率,降低误识率。
Traditionally, the recognition method for abnormal traffic has the problems of low recognition efficiency and high false positive rate. In this article, a fuzzy recognition method for abnormal traffic in transport layer of peer-to-peer overlay network based on the improved extremely randomized trees was proposed. Firstly, the information gain ratios of different features were calculated respectively and thus to obtain the feature set with low dimension. Then, the random training method was used to train the classification model. After that, the optimal classifier which was applied to the resampling data classification was used to classify the network traffic in transport layer. According to the classification result, the statistical signal model of abnormal traffic was constructed. Moreover, the signal processing method was combined with the high-order detection algorithm. Finally, the amplitude-frequency response feature was applied to the registration, so as to realize the recognition for abnormal traffic in transport layer of peer-to-peer overlay network. Simulation results show that the proposed method can effectively improve the recognition efficiency and reduce the false positive rate.
Traditionally, the recognition method for abnormal traffic has the problems of low recognition efficiency and high false positive rate. In this article, a fuzzy recognition method for abnormal traffic in transport layer of peer-to-peer overlay network based on the improved extremely randomized trees was proposed. Firstly, the information gain ratios of different features were calculated respectively and thus to obtain the feature set with low dimension. Then, the random training method was used to train the classification model. After that, the optimal classifier which was applied to the resampling data classification was used to classify the network traffic in transport layer. According to the classification result, the statistical signal model of abnormal traffic was constructed. Moreover, the signal processing method was combined with the high-order detection algorithm. Finally, the amplitude-frequency response feature was applied to the registration, so as to realize the recognition for abnormal traffic in transport layer of peer-to-peer overlay network. Simulation results show that the proposed method can effectively improve the recognition efficiency and reduce the false positive rate.
引文
[1] 李兴国,王俊峰.无线网络数据流量异常优化检测仿真研究[J].计算机仿真,2017,34(9):386-389.
[2] 于敬敬,潘景昌,孟凡龙,等.基于距离度量的LAMOST光谱中连续谱异常的自动检测[J].光谱学与光谱分析,2017,37(7):2246-2249.
[3] 陈长辉.对等网络流量信息结构异常的检测技术研究[J].应用激光,2017,40(1):134-138.
[4] 马林进,万良,马绍菊,等.基于词袋模型聚类的异常流量识别方法[J].计算机工程,2017,43(5):204-209.
[5] 燕昺昊,韩国栋,黄雅静,等.非平衡网络流量识别方法[J].计算机应用,2018,38(1):20-25.
[6] 侯爱华,高伟,汪霖.基于逻辑回归模型的流量异常检测方法研究[J].工程数学学报,2017,34(5):479-489.
[7] 应斐昊,邢宁哲,纪雨彤,等.基于KTLAD的电力数据网业务流量异常检测[J].北京邮电大学学报,2017,40(S1):108-111.
[8] 姜红红,张涛,赵新建,等.基于大数据的电力信息网络流量异常检测机制[J].电信科学,2017,33(3):134-141.
[9] 王秀娟,相从斌.基于累积量的DoS攻击检测算法[J].北京工业大学学报,2017,43(9):1328-1334.
[10] 王永恒.基于改进蚁群算法的计算机网络路由优化研究[J].电子设计工程,2017,25(20):125-128.
[11] 王景云,杨军,杨俊兴,等.基于空间自相关和概率论的土壤重金属异常值的识别方法[J].地球信息科学学报,2017,19(5):605-612.
[12] 刘建明,唐霞,李龙.基于机器学习的VoIP流量在线识别系统[J].电子技术应用,2016,42(8):133-137.
[2] 于敬敬,潘景昌,孟凡龙,等.基于距离度量的LAMOST光谱中连续谱异常的自动检测[J].光谱学与光谱分析,2017,37(7):2246-2249.
[3] 陈长辉.对等网络流量信息结构异常的检测技术研究[J].应用激光,2017,40(1):134-138.
[4] 马林进,万良,马绍菊,等.基于词袋模型聚类的异常流量识别方法[J].计算机工程,2017,43(5):204-209.
[5] 燕昺昊,韩国栋,黄雅静,等.非平衡网络流量识别方法[J].计算机应用,2018,38(1):20-25.
[6] 侯爱华,高伟,汪霖.基于逻辑回归模型的流量异常检测方法研究[J].工程数学学报,2017,34(5):479-489.
[7] 应斐昊,邢宁哲,纪雨彤,等.基于KTLAD的电力数据网业务流量异常检测[J].北京邮电大学学报,2017,40(S1):108-111.
[8] 姜红红,张涛,赵新建,等.基于大数据的电力信息网络流量异常检测机制[J].电信科学,2017,33(3):134-141.
[9] 王秀娟,相从斌.基于累积量的DoS攻击检测算法[J].北京工业大学学报,2017,43(9):1328-1334.
[10] 王永恒.基于改进蚁群算法的计算机网络路由优化研究[J].电子设计工程,2017,25(20):125-128.
[11] 王景云,杨军,杨俊兴,等.基于空间自相关和概率论的土壤重金属异常值的识别方法[J].地球信息科学学报,2017,19(5):605-612.
[12] 刘建明,唐霞,李龙.基于机器学习的VoIP流量在线识别系统[J].电子技术应用,2016,42(8):133-137.