Wireless Mesh Network Planning:A Multi-objective Optimization Approach

Djohara Benyamina; Abdelhakim Hafid; Michel Gendreau

5th International ICST Conference on Broadband Communications, Networks, and Systems

Research Article

Wireless Mesh Network Planning:A Multi-objective Optimization Approach

Cite: BibTeX Plain Text

@INPROCEEDINGS{10.1109/BROADNETS.2008.4769149,
    author={Djohara Benyamina and Abdelhakim Hafid and Michel Gendreau},
    title={Wireless Mesh Network Planning:A Multi-objective Optimization Approach},
    proceedings={5th International ICST Conference on Broadband Communications, Networks, and Systems},
    publisher={IEEE},
    proceedings_a={BROADNETS},
    year={2010},
    month={5},
    keywords={Multiobjective optimization  Planning problem  Population-based meta-heuristic search algorithm  Wireless Mesh Network},
    doi={10.1109/BROADNETS.2008.4769149}
}

Djohara Benyamina
Abdelhakim Hafid
Michel Gendreau
Year: 2010
Wireless Mesh Network Planning:A Multi-objective Optimization Approach
BROADNETS
IEEE
DOI: 10.1109/BROADNETS.2008.4769149

Djohara Benyamina¹^,*, Abdelhakim Hafid¹^,*, Michel Gendreau²^,*

1: Network Research Laboratory University of Montreal Montreal, Canada
2: CIRRELT University of Montreal Montreal, Canada

*Contact email: benyamid@iro.umontreal.ca, ahafid@iro.umontreal.ca, michel.gendreau@iro.umontreal.ca

Abstract

A modern wireless network can be neither successfully deployed nor successfully expanded without proper planning. In this paper we consider the wireless mesh network (WMN) planning problem where no much work has been done. We propose a more realistic multi-objective approach to model this problem where the two conflicting objectives of total deployment cost and network throughput are to be optimized while guaranteeing full coverage to all mesh clients. Previous contributions have mainly formulated and solved this problem by using single-objective integer linear programming formulations and exact methods. The main limitation of these approaches resides in their restriction to small sized instances. We propose a population-based meta-heuristic algorithm to solve the problem. This algorithm produces a set of good planning solutions for real-size networks thus enlarging the decision perspective of a network planner. We also discuss the effect of different parameters on the characteristics of the solutions.

Keywords: Multiobjective optimization Planning problem Population-based meta-heuristic search algorithm Wireless Mesh Network

Published: 2010-05-16
Publisher: IEEE
Modified: 2010-05-16

: http://dx.doi.org/10.1109/BROADNETS.2008.4769149