Non-exhaustive method for identification of optimal variable orderings in the decomposition of complex logic functions

Abstract

An analysis is presented to predict the input variable subsets that give efficient disjunctive decompositions of complex combinatorial logic functions. A substantially smaller number of possible solutions, compared to all possible solutions, is predicted using a comparatively simple analysis, based on a count of the covers between tuples in the function description when one or two variables are swapped across the input variable partition. From these data, prominent variables, whose positions heavily influence the function decomposition, can be predicted and optimal orderings obtained. The effectiveness and efficiency of the method is demonstrated through a number of examples.