Multi-agent based intrusion detection architecture

Intrusion detection research has been conducted for nearly 20 years but still remains in its infancy. The existing intrusion detection system architectures have a number of problems that limit their configurability, scalability, and efficiency. This paper first discusses the threats to networked computer systems, and introduces the intrusion detection mechanisms and agent technologies. On the basis of the above discussion, a multi-agent based intrusion detection architecture for LAN is proposed. This architecture can be easily extended and maintained, supporting distributed and intelligent intrusion detection. In addition, it can enhance the whole system detection efficiency greatly.