Decision-diagram-based techniques for bounded reachability checking of asynchronous systems

Bounded saturation-based CTL model checking

“…In our work we chose the one that has already proved its efficiency [19]. The bounded saturation algorithm needs additional information about traversal distance to be able to compute the reachability set below a bound.…”

“…MDDs are a highly efficient storage form for state space representation, but they do not include this information. In order to make the distance information available during the traversal, the algorithm in [19] uses EDDs instead of MDDs for storing distance measures implicitly encoded into the state space representation.…”

“…Truncating excessive states. There are two types of truncating functions in [19]. The algorithm uses these truncating functions after the traversal of the local state spaces, before finalizing the computation of nodes.…”

“…Our approach uses an enhanced version of the TruncateExact (Algorithm 6) function that differs from the one presented in [19]. We reduced the computational overhead of exploring recursively the sub-MDD by using a truncate cache (in the TruncateExact algorithm see lines 4 and 13).…”

“…Our bounded saturation algorithm extends the former one [19] with on-the-fly updates of the states and the next-state relations. This way the user does not have to provide the algorithm with the possible local state spaces of the submodels.…”