Variable Time-Frame Abstraction

Verification benefits from removing logic that is not relevant for a proof. Techniques for doing this are generally known as localization abstraction. Abstraction is often performed by selecting a subset of gates to be included in the abstracted model; the signals feeding into this subset become unconstrained cutpoints. In this paper, a novel approach to gate-level abstraction is presented, which allows for including gates in some but not all time-frames of the abstracted model. The “variable time-frame” approach is motivated by the hope of building a more scalable abstraction, because the complete logic of the abstracted model in not used in each time-frame of the unrolling. Furthermore, a similar method could make BMC and interpolation-based model checking more scalable, by considering only a subset of gates belonging to the cone-ofinfluence of the property. Preliminary results suggest that it may be a promising approach.