First Joint IEEE/IFIP Symposium on Theoretical Aspects of Software Engineering (TASE '07)
Self-adaptive Intrusion Detection System for Computational Grid
Shanghai, China
June 06-June 08
ISBN: 0-7695-2856-2
As Conventional intrusion detection systems cannot evolve with ceaselessly changing environment of computational Grid, a model of intrusion detection system named GIDIA based on immunity and multi Agents is developed with hierarchical architecture. Following definitions of immune model, detecting Agent, decision-making Agent, preventing Agent, and controlling Agent, relevant abstract mathematical models, evolving trends and inferential equations of Agents that are used in intrusion detection module, decision-making module, response module, and vaccine formation and distribution module are founded respectively. Theoretical analysis and experimental results show that GIDIA possesses better self-adaptability, higher detection rate, and offers a novel way to secure Grid member sites in the same trust community or different ones cooperatively.
Citation:
NI Jiancheng, LI Zhishu, SUN Jirong, XING Jianchuan, "Self-adaptive Intrusion Detection System for Computational Grid," tase, pp.97-106, First Joint IEEE/IFIP Symposium on Theoretical Aspects of Software Engineering (TASE '07), 2007