EPSO-SCR: Energy-Efficient PSO-Based Clustering Approach for Redundancy Reduction in WSNs

Cluster-based architectures are considered a practical approach for energy saving in wireless sensor networks. Sensor nodes are grouped in clusters based on specific criteria in order to restrict communication with the base station to a few nodes called cluster heads. These nodes aggregate data and reduce redundancy, which decreases the number of data transmissions and optimizes the network lifetime. One of the main challenges is the selection of these cluster head nodes. Our approach is based on a previous work aiming to create sub-clusters by grouping nodes based on their spatial proximity and data correlations to form similarity regions and reduce redundant data transmission. A representative node of each sub-cluster is selected to communicate with the cluster head, who, in turn, transmits data to the base station. The paper focuses on optimizing Energy saving using Particle Swarm Optimization for the sub-cluster head selection. Experimentation results show the effectiveness of our approach to extend the network lifespan while maintaining balanced energy levels on nodes.