基于集成神经网络的智能决策入侵检测系统

Intrusion Detection System Combining Misuse and Anomaly Based on Neural Network Ensemble

针对传统入侵检测系统存在误报率、漏检率较高的问题,提出了一种将误用入侵检测和异常入侵检测相结合的智能决策入侵检测系统,该系统基于集成神经网络技术,通过D-S证据理论可以将两种技术很好地结合起来,提高入侵检测系统的效率。阐述了该入侵检测系统的总体结构部署以及各组成模块的相应结构设计。

Traditional intrusion detection systems always have such problems as distortion and leakage. To solve these problems, this paper puts forward a new intrusion detection system which could combine misuse detection and anomaly detection. The system is based on neural network ensemble and it uses D-S evidence theory to combine the two intrusion detection technologies. The paper also expatiates on the main structure of the intrusion detection system and the composing module designation.