摘要
针对多层次数据中心网络容易发生流量拥塞的问题进行流量异常特征检测,以提高网络的稳定性.提出了一种基于高阶累积量后置搜索的多层次数据中心网络流量异常特征检测算法,构建多层次数据中心网络的流量传输结构模型,进行流量时频采样和时间序列分析.结合FIR滤波器进行流量抗干扰滤波预处理,利用高阶累积量的后置聚焦性,对输出的滤波数据进行高阶累积量特征提取改进和后置聚焦搜索,实现了流量序列中异常特征的准确检测和提取.仿真结果表明,采用该算法进行多层次数据中心网络流量异常检测的准确度较高,抗干扰能力较强,保障了网络的稳定和安全.
Aiming at the problem that the data center network is prone to traffic congestion,the traffic anomaly feature detection is carried out to improve the network stability.A traffic anomaly characteristics of multi level data center network cumulant Post search in detection algorithm based on traffic transmission structure model of multi level data center network,flow frequency sampling and time series analysis.Flow disturbance filtering pretreatment with FIR filter,using high order filter data of the rear focusing accumulation on the output of high order cumulants improved feature extraction and post focusing search,to achieve accurate detection of abnormal flow characteristics in the sequence extraction.The simulation results show that the algorithm has high accuracy and strong anti-interference performance,and ensures the stability and security of the network.
出处
《河南工程学院学报（自然科学版）》
2017年第1期62-66,共5页
Journal of Henan University of Engineering：Natural Science Edition
基金
国家自然科学基金(61301232)
河南省高等学校重点科研项目(17A520025)
关键词
网络
流量
检测
数据中心
高阶累积量
network
traffic
detection
data center
high order accumulation
