A new clustering algorithm based on the selection of the best energy efficient cluster head for wireless sensor networks

Wireless Sensor Networks (WSN) are the network elements which are widely used, very small size, limited battery life, data transmitting and receiving wirelessly. Because of their limited capacity, life cycles need to be extended by using as little energy as possible. In this respect, many algorithms have been developed to enable sensors to transmit or receive data with less energy. These algorithms are generally based on working of the sensors together to make the division of labor. Data transmission and reception which have a significant share in energy consumption are directly proportional with the distance between wireless sensors, bandwidth and data storage capacity. Because of this, the energy usage of data transmission of the two remote sensors will be more than the proximity sensor so the clustering method has been developed in order to reduce this distance. Therefore, Cluster Heads (KB) plays the most effective role in clustering method. In this thesis, a new clustering algorithm has been developed to minimize the energy consumed by the sensors while sending data to each other or to the base station and to extend the battery life which is vital for the sensors. This new algorithm based on an efficient clustering for Wireless Sensor Networks is called Cluster Centric-Cluster Head Selection Algorithm (Küme Merkezli- Küme Başı Seçim Algorithm (KM-KBA)). KM-KBA is a Wireless Sensor Network in each cluster is based on clustered into two clusters in itself. The first of these two clusters consists of the sensors which are located within the average distance of all the sensors and which is located in the central position of the sensors within the cluster and called the Cluster Center (CC), and the second consists of the outer cluster called sensors Outside the Cluster Center (OCC). Sensors in the Cluster Center (CC) have the priority Cluster Head (CH) in the specified period (round) for the Wireless Sensor Network's data transmission and reception. In this way, by providing the centralized sensors to be Cluster Head (CH), distance-based energy saving is provided and the life of the WSNs extended. The developed algorithm, performance evaluated with Matlab simulation tool, the number of life round of the sensors, the amount of energy spent per round and the total number of packets sent were analyzed in comparison with other algorithms.