Genetic algorithm and pure random search for exosensor distribution optimisation

Michael Poland, CD Nugent, Hui Wang, Luke Chen

Research output: Contribution to journalArticle

23 Citations (Scopus)

Abstract

The positioning, amount(s) and field of view(s) of exosensors are a fundamental characteristic of a smart home environment. Contemporary smart home sensor distribution is aligned to either: a) a total coverage approach; b) a human assessment approach. These methods for sensor arrangement are not data driven strategies, are unempirical, and frequently irrational. Little research has been conducted in relation to optimal resource allocation in smart homes environments. This study aimed to generate globally optimal sensor distributions for a smart home replica-kitchen using two distinct methodologies, namely a genetic algorithm (GA) and a pure random search algorithm (PRS), to ascertain which method is appropriate for this task. GA outperformed PRS consistently, with a coverage percentage that encapsulated an average of 43.6% more inhabitant spatial frequency data. The results of this study indicate that GA provides more optimal solutions than PRS for exosensor distributions in a smart home environment.
LanguageEnglish
Pages359-372
JournalInternational Journal of Bio-Inspired Computation
Volume4
Issue number6
DOIs
Publication statusPublished - 2012

Fingerprint

Genetic algorithms
Sensors
Kitchens
Resource allocation

Keywords

  • smart homes
  • smart environments
  • genetic algorithms
  • pure random search
  • PRS
  • optimisation
  • exosensor distribution
  • exosensors
  • sensor distribution

Cite this

@article{322d18091f02407bb2fab2e8a3355c97,
title = "Genetic algorithm and pure random search for exosensor distribution optimisation",
abstract = "The positioning, amount(s) and field of view(s) of exosensors are a fundamental characteristic of a smart home environment. Contemporary smart home sensor distribution is aligned to either: a) a total coverage approach; b) a human assessment approach. These methods for sensor arrangement are not data driven strategies, are unempirical, and frequently irrational. Little research has been conducted in relation to optimal resource allocation in smart homes environments. This study aimed to generate globally optimal sensor distributions for a smart home replica-kitchen using two distinct methodologies, namely a genetic algorithm (GA) and a pure random search algorithm (PRS), to ascertain which method is appropriate for this task. GA outperformed PRS consistently, with a coverage percentage that encapsulated an average of 43.6{\%} more inhabitant spatial frequency data. The results of this study indicate that GA provides more optimal solutions than PRS for exosensor distributions in a smart home environment.",
keywords = "smart homes, smart environments, genetic algorithms, pure random search, PRS, optimisation, exosensor distribution, exosensors, sensor distribution",
author = "Michael Poland and CD Nugent and Hui Wang and Luke Chen",
year = "2012",
doi = "10.1504/IJBIC.2012.051408",
language = "English",
volume = "4",
pages = "359--372",
journal = "International Journal of Bio-Inspired Computation",
issn = "1758-0366",
number = "6",

}

Genetic algorithm and pure random search for exosensor distribution optimisation. / Poland, Michael; Nugent, CD; Wang, Hui; Chen, Luke.

In: International Journal of Bio-Inspired Computation, Vol. 4, No. 6, 2012, p. 359-372.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Genetic algorithm and pure random search for exosensor distribution optimisation

AU - Poland, Michael

AU - Nugent, CD

AU - Wang, Hui

AU - Chen, Luke

PY - 2012

Y1 - 2012

N2 - The positioning, amount(s) and field of view(s) of exosensors are a fundamental characteristic of a smart home environment. Contemporary smart home sensor distribution is aligned to either: a) a total coverage approach; b) a human assessment approach. These methods for sensor arrangement are not data driven strategies, are unempirical, and frequently irrational. Little research has been conducted in relation to optimal resource allocation in smart homes environments. This study aimed to generate globally optimal sensor distributions for a smart home replica-kitchen using two distinct methodologies, namely a genetic algorithm (GA) and a pure random search algorithm (PRS), to ascertain which method is appropriate for this task. GA outperformed PRS consistently, with a coverage percentage that encapsulated an average of 43.6% more inhabitant spatial frequency data. The results of this study indicate that GA provides more optimal solutions than PRS for exosensor distributions in a smart home environment.

AB - The positioning, amount(s) and field of view(s) of exosensors are a fundamental characteristic of a smart home environment. Contemporary smart home sensor distribution is aligned to either: a) a total coverage approach; b) a human assessment approach. These methods for sensor arrangement are not data driven strategies, are unempirical, and frequently irrational. Little research has been conducted in relation to optimal resource allocation in smart homes environments. This study aimed to generate globally optimal sensor distributions for a smart home replica-kitchen using two distinct methodologies, namely a genetic algorithm (GA) and a pure random search algorithm (PRS), to ascertain which method is appropriate for this task. GA outperformed PRS consistently, with a coverage percentage that encapsulated an average of 43.6% more inhabitant spatial frequency data. The results of this study indicate that GA provides more optimal solutions than PRS for exosensor distributions in a smart home environment.

KW - smart homes

KW - smart environments

KW - genetic algorithms

KW - pure random search

KW - PRS

KW - optimisation

KW - exosensor distribution

KW - exosensors

KW - sensor distribution

U2 - 10.1504/IJBIC.2012.051408

DO - 10.1504/IJBIC.2012.051408

M3 - Article

VL - 4

SP - 359

EP - 372

JO - International Journal of Bio-Inspired Computation

T2 - International Journal of Bio-Inspired Computation

JF - International Journal of Bio-Inspired Computation

SN - 1758-0366

IS - 6

ER -