| Summary: | 碩士 === 國立高雄應用科技大學 === 電子工程系碩士班 === 104 === This thesis is a comparative study using metaheuristic algorithms to solve the clustering problem in Wireless Sensor Networks (WSNs). By using a distance based energy model, we evaluate the problem in term of the transmission range of sensors along with the position of the base station.
Four heuristic optimization methods are chosen due to the different characteristic in exploration and exploitation or selection and mutation process. Genetic Algorithm (GA), Particle Swarm Optimization (PSO), Differential Evolution (DE) and Grey Wolf Optimizer (GWO) are implemented, experimented in terms of comparing the performances that using the same energy model.
The metaheuristic algorithms are proved to be an excellent solution due to the outperform performance compare to LEACH. In both cases, Differential Evolution (DE) seems to be the best candidate with the first place in three of four experiments. Through implementing and doing research we had more understanding about the reason why and how heuristic optimization methods works so well through analyzing the behavior of each agent then modelling with mathematical equations. We hope that our work can be a useful reference for other researchers when applying heuristic optimization methods in clustering problem with similar energy model.
|
|---|
