An hybrid machine learning and improved social spider optimization based clustering and routing protocol for wireless sensor network

Wireless Sensor Networks (WSNs) monitor and gather environmental data by interconnecting numerous sensor nodes spread across space via wireless communication. These nodes operate on battery power, which depletes over time, thereby limiting the network’s operational lifespan. This energy constraint significantly impacts the overall longevity of the network. The primary focus of the work is to reduce energy consumption and increase the network’s lifespan. To this end, WSNs currently make extensive use of routing and clustering algorithms. This work selects an optimal Cluster Head (CH) from a set of nodes using a combination of hybrid Ant Colony Optimization (ACO) and the Improved Social Spider Cluster Optimization Algorithm (ISSOA). The selection process takes into account a number of variables, such as the nodes' residual energy, their degree and centrality, their proximity to nearby nodes, and their distance from the Base Station (BS). Furthermore, we use an optics-inspired optimization (OIO) algorithm to determine the path between the chosen CH and the BS. In order to ensure effective data transmission throughout the network, this algorithm optimizes the path based on variables including distance, node degrees, and the residual energy of nodes along the route. The simulation results show that the proposed ACO-ISSOA method significantly improves several Quality of Service (QoS) parameters. The proposed method works better than other algorithms like Low Energy Adaptive Central Hierarchical Clustering (LEACH-C), Multiple-Weight LEACH (MW-LEACH), Hybrid Genetic Algorithm with Particle Swarm Optimization (GA-PSO), and ACO-based Hierarchical Clustering (ACOHC). The ACO-ISSOA protocol improves network lifetime (5700 rounds), throughput (99%), PDR (99.5%), energy consumption (56 mJ), and execution time (45 s) for CH selection. When compared to other algorithms, the hybrid ACO-ISSO algorithm outperforms them.