Research And Implement Of Security Communication In Multi-Agent System

With the development of the Multi-Agent System research and the appearance of more MAS based application system, the security of Agent and MAS has become an important research topic. The main researches we have done in this paper are that after we analyse the possible security attacks a multi-agent system may meet, we propose a security framework, which is built upon the JADE multi-agent platform. This framework implements such security functions as identity authentication, authority controlling, message checking, event auditing. In addition we ensure the security of the data transfer with MASSF, which can prevent man-in-the-middle attack. In this paper, we first introduce the basic concepts and principles on MAS and a related international standard - FIPA, then analyzes the possible attacks which a multi agent system may meet and point out the faults of JADE. Next we describe this security framework in detail, which has effectively improved the security level of the JADE platform. At last we give a performance test for the enhanced JADE platform, and present the future researches.