Low power synthesis based on information theoretic measures

Abstract

The synthesis of circuits with reduced power consumption has grown more and more important over the last years. In this paper, we address the problem of reduction of power dissipation using dynamic power management techniques. The problem of low power synthesis corresponds to an optimal decomposition of a finite state machine reduced to choice of partitions on the set of states of prototype machine. For evaluation of the networks, the informational modeling based on entropy measure is considered. It enables one to enhance the decomposition partition search for low power synthesis.