Testing Equivalences and Fully Abstract Models for Probabilistic Processes

We present a framework in which the observable behavior of probabilistic processes is distinguished through testing. Probabilistic transition systems are used to model the operational behavior of processes. The observable behavior of processes is studied in terms of probabilities for successful interaction with tests. Based on these probabilities three equivalences are defined. We define three denotational models, and show that each model contains exactly the necessary information for verification of one of the equivalences.