NEMOCED: New Energy Model and Optimal Cluster Estimation Based on Density to Increase Lifetime in Wireless Sensor Networks

摘要

Abstract In recent years, many methods have been presented for clustering in wireless sensor networks (WSNs). Some areas have high density, and the random distribution of the nodes reduces the clustering quality. Moreover, the number of clusters is manually determined before clustering. In this paper, a new clustering algorithm called NEMOCED is presented based on the node distribution. In the NEMOCED, the best cluster head is selected according to the node distribution. Moreover, we propose a new energy model to estimate proper clusters. One of the main features of the energy model is selecting the proper clusters. It is performed based on the number of nodes and the network size. In each round, two cluster heads are selected by using the tree structure. Finally, we introduce five criteria for assessing the quality and accuracy of the proposed algorithm. The NEMOCED can perform the clustering based on the local density of nodes and choose more proper cluster heads in high-density areas. The simulation results demonstrate that the NEMOCED can significantly improve lifetime and energy consumption. Furthermore, the simulation results show that the NEMOCED algorithm has good adaptability and works well under different network lifetime definitions. All the results prove that the NEMOCED algorithm has the advantage of being suitable and efficient for large-scale WSN applications.