State assignment of finite state machines using a genetic algorithm

Abstract

The use of genetic algorithms for the generation of optimal state assignments for synchronous finite state machines (FSM) is proposed. Results are presented to show that, in all examples attempted, the resulting state assignments are better than or at least as good as those produced by SPECTRAL, NOVA and MUSTANG and also closed partition assignments. On average, the genetic algorithm produced assignments with 33% less logic than the best produced by other algorithms.

Artificial Intelligence in Logic Design

“…Although some research on the application of genetic algorithms in hardware design and electronic design automation (EDA) is reported in the literature (e.g. Almaini et al 1995;Benten and Sait 1994;Chattopadhyay and Chaurhuri 1997;Dill and Perkowski 1998;), many problems remained unsolved, and this motivated us to carry out the research reported below. As explained in Section 3, a search algorithm adequate for circuit synthesis should robustly and efficiently produce the optimal or near-optimal results.…”

A Parallel State Assignment Algorithm for Finite State Machines

“…NOVA improves on KISS and is based on a graph embedding algorithm. Genetic algorithms [11] and algorithms based on FSM decomposition [12] are also proposed. The sequential logic optimization tool SIS uses algebraic techniques to factor the logic equations.…”

State assignment for sequential circuits using multi-objective genetic algorithm

“…Generally, it is possible to find state assignments to minimise the hardware only [1][2][3]15], or the power dissipation only [4,12,16,17]. It is known, however, that minimising either the power or logic complexity could be at the expense of the other and in most cases it is not possible to find a solution that is optimum in both domains.…”

“…In [3][4][5][6], the authors proposed the use of GA to generate state assignments which minimise the gate count and/or power dissipation.…”