A High-Throughput and Arbitrary-Distribution Pattern Generator for the Constrained Random Verification

A quantum algorithm for uniform sampling of models of propositional logic based on quantum probability

“…The problem of sampling models to satisfy truth assignments of a propositional well formed formula (WFF), according to a given probability distribution belongs to the #P-complete complexity class and is harder than the nondeterministic polynomial (NP) complete class of problems. 1 This class of problems is interesting because of its wide range of applications in very large-scale integration (VLSI) functional verification via random sampling, 2 implementation of Markov logic networks, 3 and probabilistic logic based inference engines to describe uncertain situations where independent identical distribution assumptions are violated. 4 An attempt to address this class of problems by the integration of the Kolmogorovian axiomatic theory of probability and first order logic theories leads to probabilistic logic programming languages with distribution semantics.…”

On-Chip Cube-Based Constrained-Random Stimuli Generation for Post-Silicon Validation

“…For presilicon verification, the constrained-random number generator embedded in the simulator generates stimuli that satisfy userdefined constraints [30]. The constrained-random verification methodology has become the mainstream to verify systemwide functionality [38], because it eliminates a large percent of useless stimuli from traditional random verification. Hence a series of constrained-random methods are proposed to increase the quality and efficiency for verification, including pure random acceptance and rejection (A&R) [38], weighted binary decision diagram (BDD) sampling [40], interval-propagationbased sampling [12], sampling based on SAT solvers [25] and random-walk with Monte Carlo Markov Chain (MCMC) sampling [16].…”