Mobile agents in distributed computing: Network exploration

Abstract

One of the recent paradigms in networked distributed computing is the use of mobile agents. Mobile agents are software robots that can autonomously migrate from node to node within a network. Although mobile agents can be easily implemented over a message passing network, they provide an abstraction for designing algorithms in a non-traditional way which can be quite natural for certain problems, such as searching, monitoring or intruder detection. A principle sub-task in most algorithms for mobile agents is the traversal of the network. We focus on this problem of exploring an initially unknown network with one or more mobile agents. We also consider the related problem of constructing a map of the environment being explored by the mobile agents.