Computational methods of minimization of multiple functions

Abstract

The article discusses computational methods for minimizing multivalued functions. The problems of the minimization of combinational circuits are investigated. An effective heuristic method intended to simplify the canonical forms of arbitrary functions was developed. The method allows to reduce the cost necessary to build a combinational circuit. The proofs of the theorem and definitions about the operation of gluing and absorption to DNF of multivalued functions in the class of disjunctive normal forms are privatized. Examples are given for applying the gluing and absorbing operation to DNF Df , we obtain abbreviated DNF Dl .

Algebraic Recognition Approach in IoT Ecosystem

“…An algorithm operator was developed, which is considered correct for problem Z, represents the sum of q operators from the model for calculating estimates, and is described by a set of 3 • n • m • q (where n is the number of predetermined features, m is the number of reference objects, q is the set of recognized objects) numerical parameters. An operator belonging to the linear closure of a model of the type of calculation of estimates was constructed [19][20][21][22]. The completeness of the linear closure of this model was proven for all problems in which for each class there is at least one stationary pair (u, v), and this correct algorithm is written explicitly.…”