Combined Spectral Techniques for Boolean Matching

Signatures are applied as fast filters during Boolean matching in technology mapping and formal verification. They are used to efficiently find candidates of corresponding variables in pairs of Boolean functions. In this paper, we combine present techniques of signature computation with representations of Boolean functions in the spectral domains. We obtain signatures of Boolean functions and their variables, which are efficiently computable using decision diagrams, though they show high accuracy, when applied to the Actel-2 FPGA library.