Performance Evaluation of WMN-GA System in Node Placement in WMNs for Different Distributions of Mesh Clients and Different Selection and Mutation Operators

Article type: Research Article

Authors: Barolli, Admir^{1; *} | Sakamoto, Shinji² | Oda, Tetsuya² | Spaho, Evjola³ | Ikeda, Makoto⁴ | Barolli, Leonard⁴

Affiliations: [1] Logos University, Rr. Dritan Hoxha, Tirana, Albania. E-mail: [email protected] | [2] Graduate School of Engineering, Fukuoka Institute of Technology (FIT), 3-30-1 Wajiro-Higashi, Higashi-Ku, Fukuoka 811-0295, Japan. E-mails: [email protected], [email protected] | [3] Department of Electronics and Telecommunication, Polytechnic University of Tirana, Mother Tereza Square, Nr. 4, Tirana, Albania. E-mail: [email protected] | [4] Department of Information and Communication Engineering, Fukuoka Institute of Technology (FIT), 3-30-1 Wajiro-Higashi, Higashi-Ku, Fukuoka 811-0295, Japan. E-mails: [email protected], [email protected]

Correspondence: [*] Corresponding author.

Abstract: Wireless Mesh Networks (WMNs) have become an important networking infrastructure due to their low cost for providing broadband connectivity. Issues for achieving the network connectivity and user coverage are related to the node placement problem. Several optimization problems are showing their usefulness to the efficient design of WMNs. These problems are related to optimizing network connectivity, user coverage and stability. In this paper, we formulate the optimization problems using a multi-objective optimization model. For the mesh router nodes placement, the bi-objective optimization problem is obtained consisting in the maximization of the size of the giant component in the mesh routers network (for measuring network connectivity) and that of user coverage. We evaluate the performance of WMN-GA system for node placement problem in WMNs. For evaluation, we consider Normal, Exponential and Weibull Distribution of mesh clients and different selection and mutation operators. The population size is considered 64 and the number of generation 200. The simulation results show that WMN-GA system performs better for Single Mutation, Linear Ranking selection and Normal distribution of mesh clients.

Keywords: WMNs, GAs, population size, number of generations, connectivity, coverage

Journal: Informatica, vol. 27, no. 3, pp. 489-502, 2016