WMN-PSODGA - An Intelligent Hybrid Simulation System for WMNs Considering Load Balancing

Abstract

Wireless mesh networks (WMNs) are becoming an important networking infrastructure because they have many advantages, such as low cost and increased high-speed wireless Internet connectivity. In the authors' previous work, they implemented a hybrid simulation system based on particle swarm optimization (PSO) and distributed genetic algorithm (DGA), called WMN-PSODGA. Moreover, they added to the fitness function a new parameter for mesh router load balancing a number of covered mesh clients per router (NCMCpR). In this article, the authors consider Exponential, Weibull, and Normal distributions of mesh clients and carry out a comparison study. The simulation results show that the performance of the Exponential, Weibull and Normal distributions was improved by considering load balancing when using WMN-PSODGA. For the same number of mesh clients, the Normal distribution behaves better than the other distributions. This is because all mesh clients are covered by a smaller number of mesh routers and the standard deviation is improved by effectively using NCMCpR.