Improving Compositional Verification of State-based Models by Reducing Modular Unbalance

Download

PDF Download (106Kb)

Description/Abstract

Compositional Verification is a viable way to tackle the state explosion problem. However, the decomposition of a system into smaller parts is not a trivial problem, and dividing the specification into modules can be regarded as one of the main issues that concerns a compositional approach. This paper concentrates on the application of compositional verification to state-based models, in order to reduce the number of nodes assigned to memory, thus avoiding state explosion and speeding up the verification. Furthermore, we investigate and propose an estimation method that improves the compositional verification process in modular designs, such that the amount of memory required by the process is minimised. This method has been applied to a real-life embedded system, producing meaningful results without the need of data abstraction.