A Distributed and Compromise-tolerant Mobile Agent Protection Scheme

In this extended abstract paper we address the problem of protecting mobile agents from possibly malicious hosts they could visit in an open e-commerce environment, the Internet. The novel approach presented is to split up a task that otherwise could be employed by a single agent into a group of agents, which communicate and cooperate by executing on different hosts. The approach has its roots in what is known as Information Dispersal for security to provide faultand compromise-tolerance. An open distributed support environment to realize the approach is also briefly presented.