IoT based sensor network clustering for intelligent transportation system using meta-heuristic algorithm

Internet of Things (IoT) based sensor networks have been established as a pillar in intelligent communication systems for efficiently handling roadside congestion and accidents. These IoT networks sense, collect, and process data on a real-time basis. However, IoT based sensor network clustering has various energy constraints such as inefficient routing due to long-haul transmission, hot spot problem, network overhead, and unstable network whenever deployed along with the roadside that affect their architecture. In such networks, clustering techniques play a crucial role in extending the lifespan and optimizing the routes by integrating sensor devices through clusters. Therefore, a meta-heuristic algorithm for clustering in IoT sensor networks for an intelligent transportation system is proposed. In this work, the seagull optimization algorithm is applied for clustering by considering residual and average energy, node spacing, and distance fitness parameters. Moreover, this work also considers the dynamic communication range of the cluster heads for increasing the stability period and lifetime of the proposed networks. The experiment results demonstrate that the proposed Seagull optimization algorithm for clustering in IoT networks (SOAC-IoTNs) and Seagull optimization algorithm for clustering in IoT networks with dynamic communication range (SOAC-IoTNs-DR) achieve a significant increase in the stability period and network lifetime, with percentage increments of 55.68% and 71.47%, and 10.03% and 88.66% respectively, compared to the existing optimized genetic algorithm for cluster head selection with single static sink (OptiGACHS-StSS).