Premium
Implementation of cellular genetic algorithms on a CNN chip: Simulations and experimental results
Author(s) -
Horváth András,
Rásonyi Miklós
Publication year - 2012
Publication title -
international journal of circuit theory and applications
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.364
H-Index - 52
eISSN - 1097-007X
pISSN - 0098-9886
DOI - 10.1002/cta.1823
Subject(s) - computer science , compiler , exploit , travelling salesman problem , cellular neural network , genetic algorithm , parallel computing , algorithm , chip , artificial neural network , computer engineering , computer architecture , artificial intelligence , machine learning , programming language , telecommunications , computer security
SUMMARY One of the most challenging recent problems in engineering is the development of proper topographic and parallel algorithms that can exploit the high performance of multi‐parallel architectures like cellular neural networks. We have developed, simulated and implemented a cellular version of the well‐known genetic algorithm which ideally suits these devices. The paper contains not only simulations, but an actual implementation and performance analysis on an existing architecture, the Xenon _ v 3 digital cellular network, its emulator and its compiler. We support our conclusions by testing the algorithm on the traveling salesman problem. Copyright © 2012 John Wiley & Sons, Ltd.