摘要
针对传统入侵检测系统易导致系统性能瓶颈的突出问题,受高等生物免疫系统的启发,文章突破传统入侵检测系统外壳式防御的架构基础,设计了一种适用于可编程数据平面的仿生免疫入侵检测方法。该方法利用仿生固有免疫系统进行流量过滤,初步拦截部分入侵流量,对于仍然存疑的流量则启动仿生适应性免疫系统进行深度特征采集、识别与处理,实现了对入侵流量的高效检测。实验结果表明,该方法能够实现较高的检测准确率和较低的控制器负载。
This study,aiming at the prominent issue of performance bottlenecks in traditional intrusion detection systems,drawed inspiration from the higher biological immune system and broken through the architectural foundation of the shell-based defense approach in traditional intrusion detection systems.A bio-inspired immune intrusion detection method suitable for programmable data planes was designed.This method utilized the innate immune system to filter traffic,preliminarily intercepting some intrusive traffic.For traffic that remains suspicious,the bio-inspired adaptive immune system was activated to conduct deep feature collection,identification,and processing,achieving efficient detection of intrusive traffic.Experimental results demonstrate that this method can achieve high detection accuracy and low controller load.
作者
孙南
秦中元
胡爱群
李涛
SUN Nan;QIN Zhongyuan;HU Aiqun;LI Tao(School of Cyber Science and Engineering,Southeast University,Nanjing 211189,China;PLA Army Academy of Artillery and Air Defense,Nanjing 211132,China;Frontiers Science Center for Mobile Information Communication and Security,Southeast University,Nanjing 211189,China)
出处
《信息网络安全》
北大核心
2025年第8期1263-1275,共13页
Netinfo Security
基金
国家自然科学基金[U22B2026]
中央高校基本科研业务费专项资金[2242022k60005]。
关键词
仿生免疫
可编程数据平面
入侵检测
P4语言
bionic immunity
programmable data plane
intrusion detection
P4 language
