PrOCov: Probabilistic output coverage model

In order to guarantee high level of reliability of current complex digital systems, a robust functional verification process is mandatory. Random constrained functional verification has been a common technique used in the industry, but sound coverage models are needed in order to monitor and limit the amount of random testing. Item coverage refers to quantitative metrics based on occurrences of system parameters or variables, in general, specified under verification engineers expertise, particularly the output coverage modeling. In most cases, the actual output value distribution does not conform the established coverage model profile, leading to testbench execution time overhead. This work presents a methodology for a fast computation of profile similar to the real output value distribution, to assist the engineer in the selection of the proper check points or output ranges of interest. At the core of this methodology is the Probabilistic Output Coverage (PrOCov) tool, which was developed with the above goals.