Design and implementation of self-protection agent for network-based intrusion detection system

Abstract

Static secure techniques, such as firewall, hierarchy filtering, distributed disposing, layer management, autonomy agent, secure communication, were introduced in distributed intrusion detection. The self-protection agents were designed, which have the distributed architecture, cooperate with the agents in intrusion detection in a loose-coupled manner, protect the security of intrusion detection system, and respond to the intrusion actively. A prototype self-protection agent was implemented by using the packet filter in operation system kernel. The results show that all the hosts with the part of network-based intrusion detection system and the whole intrusion detection system are invisible from the outside and network scanning, and cannot apperceive the existence of network-based intrusion detection system. The communication between every part is secure. In the low layer, the packet streams are controlled to avoid the buffer leaks existing in some system service process and back-door programs, so as to prevent users from misusing and vicious attack like Trojan Horse effectively.