Evolution of digital circuits using CUDA to determine the fitness function in genetic algorithm

Título: Evolution of digital circuits using CUDA to determine the fitness function in genetic algorithm

Autores: Lacerda, Wilian Soares; Peixoto, Luiz Henrique Rowan; Oliveira, Thomaz Chaves de A.

Resumo: This paper proposes the use of Evolutionary Computing applied to the synthesis of digital circuits in the disjunctive normal form. Due to the computational cost of the employed technique, the platform of digital synthesis was implemented using parallel processing in order to reduce processing time. Performance comparisons were accomplished, proving that the digital platform using CUDA had considerable overall gain performance. A case study of digital synthesis was accomplished with simple digital comparators circuits. The presented result in this paper demonstrates the feasibility of the synthesis of combinational digital circuits by Evolutionary Computation, which represents a new design methodology for electronic circuits.

Palavras-chave:

Páginas: 8

Código DOI: 10.21528/CBIC2013-091

Artigo em pdf: bricsccicbic2013_submission_91.pdf

Arquivo BibTex: bricsccicbic2013_submission_91.bib