Optimizing Wireless Sensor Network Lifespan Through Weighted Fitness Function in K-Means Clustering and Cluster Head Selection

One of the primary challenges in Wireless Sensor Networks (WSNs) is the efficient management of energy con sumption and maintaining robust network connectivity over time. To address this issue, we propose a model that initially employs k-means clustering to form clusters and subsequently selects cluster heads in each round based on three critical factors: remaining energy, distance to the base station, and distance to the cluster center. By leveraging k-means clustering at the beginning, optimal cluster formation is achieved, ensuring balanced initial energy distribution. In subsequent rounds, dynamic selection of cluster heads is performed using weighted weights for each parameter to enhance network performance. Various weight values were tested to determine the most effective configuration. Our approach effectively balances energy consumption among nodes, extends network lifetime, and maintains strong connectivity, thereby addressing key challenges in the deployment and operation of WSNs.