摘要
为实现对网络数据流中异常的完全检测,提高网络数据流的安全性与可靠性,引进数学模型,从硬件与软件两个方面,设计一种针对网络数据流的异常检测系统。首先,选择HTLMK-15000型号的采集设备与F1501-G1400型号的单片机作为系统的主要硬件。其次,在硬件设备的支撑下,获取网络数据流的异常特征与属性,引进分段线性值函数(Piecewise Linear Value Function,PLVF)数学模型,对采样中的局部异常点进行强化训练,并通过对网络数据流尺度的分解,实现对数据流异常的识别与检测。最后,选择基于改进单类支持向量机(One Class Support Vector Machine,OCSVM)技术的异常检测系统作为传统系统,开展对比实验。实验结果表明,设计的系统在实际应用中可以实现对样本中全部异常点的精准检测,检测结果更加全面。
In order to realize the complete detection of anomalies in network data flow and improve the security and reliability of network data flow,a mathematical model is introduced to design an anomaly detection system for network data flow from both hardware and software aspects.Firstly,select HTLMK-15000 model acquisition equipment and F1501-G1400 model microcontroller as the main hardware of the system.Secondly,With the support of hardware equipment,the abnormal characteristics and attributes of network data flow are obtained,the Piecewise Linear Value Function(PLVF) mathematical model is introduced,and the local abnormal points in the sampling are intensively trained.detection.Finally,the anomaly detection system based on the improved One Class Support Vector Machine(OCSVM) technology is selected as the traditional system,and comparative experiments are carried out.The experimental results show that the designed system can accurately detect all abnormal points in the sample in practical applications,and the detection results are more comprehensive.
作者
时文俊
SHI Wenjun(Department of Basic Courses,Zhengzhou Shengda University,Zhengzhou Henan 451191,China)
出处
《信息与电脑》
2022年第18期115-117,共3页
Information & Computer
关键词
数学模型
数据流异常
数据流尺度
mathematical model
abnormal data flow
data flow scale
作者简介
时文俊(1980-),女,河南新乡人,硕士研究生,教授。研究方向:应用数学及数学教育。
