一种基于预测模型的网络安全风险实时预测方法设计

详细信息    查看全文 | 推荐本文 |

英文篇名：A Real-time Prediction Method of Network Security Risk Based on Predictive Model 作者：李鑫 英文作者：LI Xin;Department of Information Engineering,Zhenzhou Shengda University of Economics,Business & Management; 关键词：网络安全 ; 安全风险预测 ; 小波神经网络 ; 网络攻击 ; Morlet ; 动态预测 英文关键词：network security;;security risk prediction;;wavelet neural network;;network attacks;;Morlet;;dynamic prediction 中文刊名：CGGL 英文刊名：Journal of Chongqing University of Technology(Natural Science) 机构：郑州升达经贸管理学院信息工程学院; 出版日期：2019-02-15 出版单位：重庆理工大学学报(自然科学) 年：2019 期：v.33;No.398 基金：河南省科技厅2018年度重点研发与推广项目(182102210139,182102110277) 语种：中文; 页：CGGL201902022 页数：6 CN：02 ISSN：50-1205/T 分类号：138-143

摘要

精确的网络安全风险预测能够动态地降低网络入侵的风险,还有助于提高网络稳定运行的效率。作为信息安全的一个关键组成要素,传统的网络安全风险预测对网络资源的风险值评估是静态的,存在安全风险动态预测偏差较大的问题。因此,为了提高网络安全风险实时预测的准确性,提出了一种基于预测模型的网络安全风险实时预测方法。该方法首先根据网络攻击序列和安全形势评估构建预测模型,然后将小波神经网络预测算法应用于攻击强度观测序列的数据分析,并选取Morlet小波函数作为激励函数。仿真实验结果显示:相比其他预测方法,提出方法具有更高的网络安全风险预测准确性。
        Precise cyber security risk prediction can dynamically reduce the risk of network intrusion and also help improve the efficiency of the network 's stable operation. As a key component of information security,the traditional network security risk prediction is static to the evaluation of network resource risk,and there is a problem that the dynamic risk of security risk is large.Therefore,in order to improve the accuracy of real-time network security risk prediction,a real-time network security risk prediction method based on prediction model is proposed. The method first constructed a prediction model based on the network attack sequence and security situation assessment. Then the wavelet neural network prediction algorithm was applied to the data analysis of the attack intensity observation sequence,and the Morlet wavelet function was selected as the excitation function. The simulation experimental results show that compared with other prediction methods,the proposed method has higher accuracy of network security risk prediction.

引文

[1]黄光球,李艳.基于粗糙图的网络风险评估模型[J].计算机应用,2010,30(1):190-195.
    [2] CURIAC D I. Wireless Sensor Network Security En-hancement Using Directional Antennas:State of the Artand Research Challenges[J]. Sensors,2016,16(4):488.
    [3]唐晓清,刘莹,刘念祖.含根网络的可靠性与稳定性研究[J].重庆师范大学学报(自然科学版),2017,34(5):65-71.
    [4] LIYANAGE M,ABRO A B,YLIANTTILA M,et al. Op-portunities and Challenges of Software-Defined MobileNetworks in Network Security[J]. IEEE Security&Pri-vacy,2016,14(4):34-44.
    [5] REN W,JIANG X H,SUN T F. RBFNN-based Predictionof Networks Security Situation[J]. Computer Engineering&Applications,2006,42(31):136-138.
    [6] SHI B,XIE X Q. Research on network security situationforecast method based on D-S evidence theory[J]. Com-puter Engineering&Design,2013,34(3):821-825.
    [7] SHI Y,LI R,YU Z,et al. An immunity-based time seriesprediction approach and its application for network secur-ity situation[J]. Intelligent Service Robotics,2015,8(1):1-22.
    [8]苏银皎,苏铁熊,王大振,等.改进小波神经网络用于火电厂污染物排放量的预测[J].计算机科学,2016,43(s1):508-511.
    [9] LUKASIEWYCZ M,LUKASIEWYCZ M,LUKASIEW-YCZ M,et al. Open source model and simulator for real-time performance analysis of automotive network security[J]. Acm Sigbed Review,2016,13(3):8-13.
    [10]郭祖华,李扬波,徐立新,等.面向云计算的网络安全风险预测模型的研究[J].计算机应用研究,2015(11):3421-3425.
    [11] MOREIRA R,MORENO R,STRBAC G. Value of correc-tive network security for distributed energy storage appli-cations[J]. Iet Generation Transmission&Distribution,2016,10(7):1758-1767.
    [12] YANG H J,HU X. Wavelet neural network with improvedgenetic algorithm for traffic flow time series prediction[J]. Optik-International Journal for Light and ElectronOptics,2016,127(19):8103-8110.
    [13] MAN D,WANG Y,WU Y,et al. A Combined PredictionMethod for Network Security Situation[C]//InternationalConference on Computational Intelligence and SoftwareEngineering. IEEE,2010:1-4.
    [14]张磊,孙长青.高阶残差修正GM(1,1)区间预测模型及其应用[J].兵器装备工程学报,2017(2):185-189.
    [15] YU Q,LYU J,JIANG L,et al. Traffic Anomaly DetectionAlgorithm for Wireless Sensor Networks Based on Im-proved Exploitation of the GM(1,1)Model[J]. Interna-tional Journal of Distributed Sensor Networks,2016,12(7):256-267.