Constrained Cartesian Genetic Programming - A New Paradigm for Evolving Imprecise Multipliers

Abstract

Optimization of multiplier architecture at the gate level is essential to reduce the silicon area consumed by the hardware for various low power applications. Further the conventional optimization algorithms like Cartesian Genetic Programming (CGP) is more suitable for evolving combinational circuits both at the gate-level and at the functional-level. We propose Constrained Cartesian Genetic Programming (CCGP), a variant of CGP to evolve various architectures of digital arithmetic circuits, especially multipliers. The proposed methodology CCGP is more suitable for evolving both imprecise and precise multipliers and has got various performance benefits over the other existing optimization methods. In this paper we make an exhaustive approach for the imprecise multiplier problem and we analyze the CCGP for its performances
Copyright © 2014 Praise Worthy Prize - All rights reserved.