Augmenting Energy Sustainability of Static Nodes Using Hybrid KGNN-AHP Driven Approach for IoT-Based Heterogeneous WSN

As the nodes used in Internet of Things (IoT)- based wireless sensor network (WSN) are constrained by the limited source of energy, contemporary applications incorporate the heterogeneous energy model WSN. Although the node energies are heterogeneous in application, further study is required to improve the potential of heterogeneous WSN and not much is explored on Graph Neural Network (GNN) clustering and Routing methods, which impels this study on novel optimal clustering of nodes with efficient cluster head selection and routing for heterogeneous WSN. In the proposed work, a novel clustering model with K-nearest neighbour (KNN) and GNN clustering, followed by the Analytic Hierarchical Process (AHP) weight-based cluster head (CH) selection method to find the optimal and eligible CH with classified fitness functions. The chosen parameters for the fitness functions are the base station to node distance, lifetime of the nodes in the network, average number of neighbouring nodes in the cluster, peak power transmission by the node in the network, and the average lifetime of nodes in the cluster. Additionally, the proposed algorithm ensures the eligibility of the optimal relay nodes using the GNNSage based routing and steady-state of data transmission from source nodes to the base station. As the model is designed for static nodes deployment for monitoring environmental changes in realtime application. The simulation results exhibit the enhancement of network lifetime and data transmission by following the proposed algorithm for static network.