A comparison study for different settings of crossover and mutation rates using WMN-GA simulation system

Abstract

In this paper, we propose and implement a system based on Genetic Algorithms (GAs) called WMN-GA. We evaluate the performance of WMN-GA for 0.7 crossover rate and 0.3 mutation rate, Exponential Ranking and different distribution of clients considering size of giant component parameter. The simulation results show that for Normal Distribution the system has better performance. We carried out also simulations for Normal Distribution and 0.8 crossover rate and 0.2 mutation rate. The simulation results shows that the setting for 0.7 crossover rate and 0.3 mutation rate offers better connectivity. © 2012 Springer Science+Business Media B.V.