Section: New Results

Automata for Data Words

We studied data words, i.e, strings where each position carries both a label from a finite alphabet and some values from an infinite domain. Data words are suitable to model the behavior of concurrent systems with dynamic process creation, as the infinite alphabet can be used to represent an unbounded number of process identifiers. A variety of formalisms, including logic and automata, have been studied to specify sets of data words in the context of verification. However, logic and automata that capture dynamic communicating systems were missing. We closed this gap and developed a quite general logical and automata-theoretic framework for the specification and implementation of sets of data words. On the specification side, we considered a fragment of monadic second-order (MSO) logic, which comes with a predicate to test two word positions for data equality. As a model of an implementation, we introduced class register automata. Our model combines the well known models of register automata and class memory automata, and it indeed captures dynamic communicating automata, whose semantics can be described as a set of message sequence charts. We studied the realizability problem and show that every formula from the existential fragment of MSO logic can be effectively translated into a class register automaton. These results were presented at CONCUR'11 [49] .