Anonymous Scheme for Secure Mobile Agent Migration Using Mignotte's Sequence and Back Propagation Artificial Neural Networks

Abstract

A mobile agent is an autonomous executing small piece of program that can relocate from one host to another in a non-homogeneous network under its own control. Mobile agents are designed to execute certain assigned tasks by the owner. In the life cycle of mobile agents, these pass over many hosts for the execution of tasks. Mobile agent’s scheme is widely used in distributed computing because of its dynamic nature, less bandwidth and less computation power. Execution of mobile agents exploits codes; data and state upraise the security issues. Malicious agents can attack mobile agents during the transmission and at the time of execution on the host because mobile agents carry delicate information of owners. The protection of mobile agents and platforms is a sensitive issue. This article focuses on the security issue of mobile agents and platforms. Provide the anonymous secure framework for mobile agent and platform security by using optimize secret key management based on Mignotte's sequence and artificial neural network. In an anonymous secret sharing technique, secret keys regenerated without knowledge of which mobile agents hold which share. That is, in such a technique the secret can be reconstructed from the shares without the identities of mobile agents.