Parallel evolutionary algorithms in telecommunications: two case studies

Sequential and parallel evolutionary algorithms (EAs) are developed and evaluated on two hard optimisation problems arising in the field of Telecommunications: designing error correcting codes, and finding optimal placements for antennas in radio networks. Different EA models (generational, steadyst...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Alba, Eladio, Cotta, C., Chicano, F., Nebro, A.J.
Formato: Objeto de conferencia
Lenguaje:Inglés
Publicado: 2002
Materias:
Ciencias Informáticas
Parallel
Algorithms
ARTIFICIAL INTELLIGENCE
Parallel Evolutionary Algorithms
Telecommunications
Error Correcting Codes
Radio Network Design
Efficiency
Acceso en línea:http://sedici.unlp.edu.ar/handle/10915/22930
Aporte de:
SEDICI (UNLP) de Universidad Nacional de La Plata
id I19-R120-10915-22930
record_format dspace
institution Universidad Nacional de La Plata
institution_str I-19
repository_str R-120
collection SEDICI (UNLP)
language Inglés
topic Ciencias Informáticas
Parallel
Algorithms
ARTIFICIAL INTELLIGENCE
Parallel Evolutionary Algorithms
Telecommunications
Error Correcting Codes
Radio Network Design
Efficiency
spellingShingle Ciencias Informáticas
Parallel
Algorithms
ARTIFICIAL INTELLIGENCE
Parallel Evolutionary Algorithms
Telecommunications
Error Correcting Codes
Radio Network Design
Efficiency
Alba, Eladio
Cotta, C.
Chicano, F.
Nebro, A.J.
Parallel evolutionary algorithms in telecommunications: two case studies
topic_facet Ciencias Informáticas
Parallel
Algorithms
ARTIFICIAL INTELLIGENCE
Parallel Evolutionary Algorithms
Telecommunications
Error Correcting Codes
Radio Network Design
Efficiency
description Sequential and parallel evolutionary algorithms (EAs) are developed and evaluated on two hard optimisation problems arising in the field of Telecommunications: designing error correcting codes, and finding optimal placements for antennas in radio networks. Different EA models (generational, steadystate and cellular) are compared on these two problems, both in sequential and parallel versions. We conclude that the cellular EA is a very effective technique, consistently finding the optimum, although it is slower than a steady-state EA. A distributed steady-state EA is shown to be the best approach, achieving the same success rate than the cellular EA in much lower time. Furthermore, it is shown that linear speedups are possible when using separate processors.
format Objeto de conferencia
Objeto de conferencia
author Alba, Eladio
Cotta, C.
Chicano, F.
Nebro, A.J.
author_facet Alba, Eladio
Cotta, C.
Chicano, F.
Nebro, A.J.
author_sort Alba, Eladio
title Parallel evolutionary algorithms in telecommunications: two case studies
title_short Parallel evolutionary algorithms in telecommunications: two case studies
title_full Parallel evolutionary algorithms in telecommunications: two case studies
title_fullStr Parallel evolutionary algorithms in telecommunications: two case studies
title_full_unstemmed Parallel evolutionary algorithms in telecommunications: two case studies
title_sort parallel evolutionary algorithms in telecommunications: two case studies
publishDate 2002
url http://sedici.unlp.edu.ar/handle/10915/22930
work_keys_str_mv AT albaeladio parallelevolutionaryalgorithmsintelecommunicationstwocasestudies
AT cottac parallelevolutionaryalgorithmsintelecommunicationstwocasestudies
AT chicanof parallelevolutionaryalgorithmsintelecommunicationstwocasestudies
AT nebroaj parallelevolutionaryalgorithmsintelecommunicationstwocasestudies
bdutipo_str Repositorios
_version_ 1764820467916472323

Ejemplares similares