An Energy-Efficient Clustering Model for Wireless Sensor Networks Using Modified K-Means Algorithm

Abstract

Wireless Sensor Networks (WSNs) are becoming essential for many applications, ranging from smart cities to environmental monitoring. WSNs comprises a collection of deployed sensor nodes to execute specified objectives in a certain area. Since batteries can only hold so much energy, one of the most crucial topics of research is how to use energy efficiently in order to extend the lifespan of sensors. One of the most popular methods for lowering energy consumption is clustering, and clustering routing protocols are methods for preserving energy to increase the lifetime of a wireless sensor network. The K-Means algorithm is one of the clustering techniques that requires prior knowledge of the clusters. This study proposes a mathematical model to determine the optimal number of clusters in WSNs, reducing energy consumption by up to 97%. Choosing the number of clusters at random could use more energy and reduce the network lifetime. This paper seeks to present a new approach for determining out the optimal number of clusters in a WSN. The proposed model tests the WSN performance by using a mathematical model and implementing it as a simulation technique in MATLAB. It considers the key WSN characteristics, including the deployed area size (100 × 100), the number of rounds (100, 200, 300, 400, and 500), and the number of sensor nodes (500). This study demonstrated that our revised approach to selecting the number of sensor network clusters reduced overall energy consumption by 97% when compared to the conventional model, hence increasing the networks' overall lifespan.