Abstract: Wireless Sensor Networks (WSNs) is partitioned into clusters to save energy, advance network scalability and resourceful routing. Residual energy, cluster node distance that is related to a node with respect of its structural position in the network is used for electing cluster heads. However for choosing cluster head the optimal numbers of nodes that may belong to a cluster are not taken into consideration. The latest centrality metric "Cluster Optimal Degree Centrality"(CODC) proposed in our paper addresses residual energy of individual nodes, distance between the potential cluster head and particular member nodes to ensure better cluster head selection and cluster quality. Finally based upon the distinct centrality metric, the Fuzzy Inference System based cluster head selection method has been proposed, which takes input as Expected Residual Energy and CODC. Though the clustering can improve QoS in wireless sensor networks, the proposed method can effectively prolong the network lifetime, since lifetime is directly related with the energy of the nodes optimizing this energy consumption is very important. The simulation results show that the proposed method performs better than LEACH resulting in high throughput and QoS.

Keywords: Cluster head selection; cluster optimal degree centrality; energy efficiency; fuzzy inference.