为了探索网络功能虚拟化(Network Function Virtualization,NFV)环境下的网络安全技术,解决网络异常检测及定位问题,文章通过采用矩阵差分分解,着眼于提升网络异常情况下的检测精确度与定位,在构建的NFV网络模型中利用不同强度的异常流场景,深入分析了网络异常对系统性能的影响,测试了基于矩阵差分分解的MADEL(Matrix Analysis for Detection and Location)算法在不同场景下的表现.研究结果表明,MADEL算法能够有效适应不同异常环境,随着异常流强度的增加,算法的检测与定位效果为NFV环境下的网络安全管理提供了有力的技术支持.
Research on network security technology supported by network function virtualization
This study aims to explore network security technologies in the context of NFV(Network Function Virtualization)and address issues related to network anomaly detection and localization.In the study,matrix differential decomposition is used to improve the detection accuracy and localization of network anomalies.In the constructed NFV network model,the different intensity anomaly flow scenarios are utilized to deeply analyze the impact of network anomalies on system performance.The performance of the MADEL(Matrix Analysis for Detection and Location)algorithm based on matrix differential decomposition is tested in different scenarios.The research results indicate that the MADEL algorithm can effectively adapt to different abnormal environments.As the intensity of abnormal flow increases,the detection and localization performance of the algorithm provides strong technical support for network security management in NFV environments.
NETL
NSTL
万方数据
