NIDS Research Based on Artificial Immunology

摘要

Current network intrusion detection systems are of low intelligence level and have the main deficiency as being unable to detect new intrusive behaviors of unknown signatures. The protection mechanism of natural immune system has brought us inspirations to design a novel network intrusion detection system. The research on modeling a NIDS with natural immune system just started, including the negative selection algorithm proposed by S. Forrest and the basic system model proposed by J. Kim. Based on their works, this paper proposed a novel system structure including affinity mutation, which was used to improve the performance of anomaly detection, and established an basic system based on artificial immunology. This paper stressed on the novel construction and testing experiments. Result of the experiments proved that the application of the protection mechanism of natural immune system to network intrusion detection system has an exciting perspective.