Distributed security algorithms by Mobile Agents

Mobile Agents have been extensively studied for several years by researchers in Artificial Intelligence and in Software Engineering. They offer a simple and natural way to describe distributed settings where mobility is inherent, and an explicit and direct way to describe the entities of those settings, such as mobile code, software agents, viruses, robots, web crawlers, etc. Further, they allow to express immediately notions such as selfish behaviour, negotiation, cooperation, etc arising in the new computing environments. As a programming paradigm, they allow a new philosophy of protocol and software design, bound to have an impact as strong as that caused by that of object-oriented programming. As a computational paradigm, mobile agents systems are an immediate and natural extension of the traditional message-passing settings studied in distributed computing. In spite of all this, mobile agents systems have been largely ignored by the mainstream distributed computing community. It is only in the last few years that several researchers, some motivated by long investigated and well established problems in automata theory, computational complexity, and graph theory, have started to systematically explore this new and exciting distributed computational universe. In this paper we describe some interesting problems and solution techniques developed in this investigations.