Hybrid Based Selective Genetic Algorithm

Abstract

Wireless Sensor Network WSN deployment is an active research area. Its goal is to deploy sensors in certain environment efficiently to optimize some evaluation measures. Meta heuristic searching optimization approaches have been proven to be effective in solving WSN deployment problem. They have strong power in exploring the solution space and converging toward theoptimal region. One key factor in achieving more exploring power in the meta-heuristic searching is the selection criteria of the elite solutions from one iteration to another. Two common selection criteria are roulette wheel and pairwise tournament. In this article, a hybrid based selection is applied under genetic algorithm for solving the problem of WSND. The hybrid based selection selects between roulette wheel and pairwise tournament in order to maintain good exploration and fast convergence toward the best solutions