M.A. Thesis:

LEACH algorithm improvement using genetic algorithm

We will work on clustering and choosing the proper cluster-head in the wireless sensor network. In symmetrical clustering methods the network is divided into several equal sections and each section will take a cluster-head regardless of the nodes in it. But in our method, which is asymmetrical clustering, using the immediate operator, the amount of centrality of nodes for the genetic algorithm is calculated and the number of the cluster-heads is chosen variably and dynamically in each round, with two other criteria of diversity and energy. In most methods the cluster-head is chosen either in distributed form, which increases energy consumption, or in focused form, which may become problematic, because one node decides for the whole network and traffic will be too heavy on that node. If that node faces a problem, the whole network will become problematic. The method suggested in the present study could decrease energy consumption by 54 percent with the use of genetic algorithm, compared to Leach algorithm.

Keywords: Lich algorithm - Genetic Algorithm - Wireless Sensor Network