Negative Selection and Niching by an Artificial Immune System for Network Intrusion Detection

This paper presents a negative selection algorithm with niching by an artificial immune system, for network intrusion detection. The paper starts by introducing the advantages of negative selection algorithm as a novel distributed anomaly detection approach for the development of a network intrusion detection system. After discussing the problems of existing approaches using negative selection for network intrusion detection, this paper presents a modified negative selection algorithm with niching, which shows diversity, generality and requires less computation time. The network packet data used in this work is then introduced and a novel genotype encoding scheme to handle this data and a corresponding fitness function is explained.