A Reinforcement in Detecting Anomalous Agents in Mobile Agent System
Qiu Yu-hui · Journal of Southwest China Normal University · 2004
Despite many benefits that mobile Agents can bring out, its users face security problem that is from malicious Agents and hosts. Though there exist some measures to protect mobile Agents from some attacks to some degree, mobile Agents could still be assaulted illegally while they are roaming from one host to another. In order to address this problem further, this paper improves an anomaly detection model for detecting malicious Agents. In particular, we focus on mobile Agent's activities by measuring its movement pattern and residence time on hosts in the original step. Moreover, we classify the normal Agent tracks and predict the next Agent's target. Finally we do some experiments to evaluate the prediction algorithm and analyze the results.