LBAS: Load-Balancing Aware Clustering Scheme for IoT-Based Heterogeneous Wireless Sensor Networks

Wireless sensor networks (WSNs) face significant challenges due to their constrained resources, particularly energy, necessitating efficient utilization to prolong their lifespan. This article introduces load-balanced adaptive clustering scheme (LBAS), a novel approach that leverages the hierarchical network structure offered by star graph-based clustering to extend the working duration of heterogeneous WSNs. Our scheme considers local information parameters of nodes as criteria for selecting cluster heads (CHs) and forming clusters. LBAS incorporates load-balancing awareness into cluster design, ensures appropriate CH selection and cluster formation, and includes a dynamic load balancing and reconfiguration phase for subsequent rounds. By utilizing the softmax function for CH election and introducing novel procedures based on graph theory concepts, the algorithm achieves efficient resource allocation and management, classifying nodes into three load-based classes. Transformation of the load-balancing process into linear equations further enables adaptive network operation. Through extensive simulations, we demonstrate the effectiveness of our LBAS scheme in prolonging the network’s lifetime compared to other clustering algorithms. These findings emphasize the potential of our approach in extending the operational duration of WSNs while ensuring efficient resource utilization.