The general aim of this research is to improve our understanding of concurrent systems and of the mathematical techniques that are useful for reasoning about them. In this specific project, interest is networks of concurrently executing, communicating processes, and especially in "dataflow-like" networks, in which processes communicate by transmitting "data values" over "channels." They are modeled using concurrent transition systems, a generalization of ordinary nondeterministic transition systems into which concurrency information has been incorporated in a convenient fashion. The goal is to classify such networks by relating their "transition structure" to their input/output behavior. This entails search for classes of networks, defined by axioms imposed on their structure as transition systems, for which one can obtain simple characterizations of the corresponding classes of input/output relations.
