SYMBOLIC MINIMIZATION OF LOGIC FUNCTIONS.

Abstract

Symbolic minimization consists of determining a two-level sum-of-products representation of a switching function in a minimal number of product terms, independently of the Boolean encoding of all (or part of) the inputs and outputs. When the objective is to determine a binary-valued description, the minimal symbolic representation determines a set of constraints for the encoding of the input and outputs into Boolean variables. Symbolic minimization is related, but not equivalent, to multiple-valued logic minimization. It represents a new concept in switching theory. Formal definitions and properties of a symbolic representation of a switching function are presented. An algorithm for symbolic minimization is described. The algorithm is implemented in a computer program called CAPPUCCINO, which is built on top of a modified version of the heuristic logic minimizer ESPRESSO-II.