Implementation and performance evaluation of a protocol for detecting suspicious hosts

Abstract

Mobile agents are software entities that consist of code, data and state, and that can migrate autonomously from host to host performing some actions on behalf of a user. Security issues restrict the use of code mobility despite its benefits. The protection of mobile agents from the attacks of malicious hosts is considered by far the most difficult security problem to solve in mobile agent systems. In this paper the authors present the implementation and the performance evaluation of a new protocol for detecting suspicious hosts in the mobile agent environment. The protocol is based on limiting the execution time of mobile agents in the executing hosts and was previously presented in [6]. The authors also present how this protocol can improve the usability of the cryptographic traces approach [12]. © Springer-Verlag Berlin Heidelberg 2003.