Paper Info
Title
Assuming multiobjective metaheuristics to solve a three-objective optimisation problem for Relay Node deployment in Wireless Sensor Networks
Abstract
Graphical abstractDisplay Omitted This paper deals with how to efficiently deploy energy-harvesting Relay Nodes in previously established low-cost static Wireless Sensor Networks (WSNs), assuming a single-tiered network model. The purpose is to optimise three conflicting objectives: Average Energy Cost, Average Sensitivity Area, and Network Reliability. This is the so-called Relay Node Placement Problem (RNPP), which is an NP-hard optimisation problem. We find many works assuming heuristics in the current literature. However, it is not the case for metaheuristics, which usually provide good results solving such complex problems. This situation led us to consider a wide range of MultiObjective (MO) metaheuristics: the two standard Genetic Algorithms NSGA-II and SPEA2, the trajectory algorithm MO-VNS, the algorithm based on decomposition MOEA/D, and two novel swarm intelligence algorithms MO-ABC and MO-FA, which are based on the behaviour of honey bees and fireflies, respectively. These metaheuristics are applied to optimise a freely available data set. The results obtained are analysed considering two MO metrics: hypervolume and set coverage. Through a widely accepted statistical methodology, we conclude that MO-FA provides the best performance on average. We also study the efficiency of this approach, verifying that it is a good strategy to optimise such networks, including some limitations. Finally, we compare this proposal to another author approach, which assumes a heuristic.
Year
DOI
Venue
2015
10.1016/j.asoc.2015.01.051
Appl. Soft Comput.
Keywords
Field
DocType
metaheuristics,coverage,optimization,reliability,wireless sensor networks,energy efficiency
Mathematical optimization,Heuristic,Swarm intelligence,Heuristics,Artificial intelligence,Reliability (computer networking),Wireless sensor network,Machine learning,Network model,Genetic algorithm,Mathematics,Metaheuristic
Journal
Volume
Issue
ISSN
30
C
1568-4946
Citations 
PageRank 
References 
7
0.43
44
Authors
2
Name
Order
Citations
PageRank
José Manuel Lanza-Gutiérrez1719.31
Juan Antonio Gómez-Pulido233443.02