Research On Mobile Agent Based Trust Management System

With the development of Internet,various things are connected,and the network security becomes increasingly serious,especially the attacks to routers.For the current rapid development of the high distributed network,such as the Internet,Ad Hoc network,sensor network and so on,The traditional cryptography based authentication mechanism has become inefficient for solving the routing security problem.To solve the above problem,trust management is introduced to monitor dynamic behaviors,evaluate trustworthiness and make decisions based on trust.Despite these advantages,as an novel security paradigm,trust based routing protocols for large-scale distributed networks also have many problems,including slow convergence of trust,high overhead,difficulties in discovering trustworthy paths.In particular,the difficulties in the collection and transmission of trust information hinder the wide application of trust management.To solve these problems,we propose a mobile-agent based trust management system.The system provides a secure and efficient platform that implements the trust information collection and trust value transfer for routing mechanism based on trust.Firstly,we conduct a security analysis of the platform,transmission and agent of mobile agent systems.On this basis,we propose a method to enhance the security of mobile agent systems.RSA public key cipher algorithm,AES and SHA-3 hash algorithm are employed to enhance the security of mobile agent systems,such as data confidentiality,authentication,etc.Based on the security enhanced mobile agent system,a trust value propagation model is proposed.By researching on the abstraction of network topology and the traversal of graphs,we discuss on the depth-first traversal and breadth-first traversal,and then design two methods for topology traversal:minimum spanning tree and flooding.Both of the two methods are able to exploit the characteristics of autonomy and mobility of mobile agents,which make the collection and distribution of the trust value more faster and flexible.Furthermore,we also defined two important agents:router defend agent and trust value transmission agent.Their communication policy and functions are detailed.Finally,we implement a experimental system to verify the proposed model.The mobile agent system IBM Aglets is employed and deployed on five PCs which emulate routers.Prim algorithm is used to calculate the minimum spanning tree.Experimental results show that the proposed scheme performs better on accelerating trust convergence and decreasing communication overhead compared with traditional solutions.In addition,the confidentiality and integrity of trust information can be assured during propagation.