A cluster-based routing in WSN for smart city applications using neural networks

Abstract

A smart city is a phenomenon that combines information technology with physical and social infrastructure to regulate a city’s cooperative intelligence. Wireless sensor networks (WSN) are the fundamental technology that smart cities use to administer and sustain their service offerings. To decrease the network’s energy consumption, clustering and multihop routing algorithms have been suggested, verified, and put into practice in the literature. This inspiration led to the development of the “energy-aware clustered route approach” in the current study, which is suggested for WSNs in smart cities. The presented method focuses on choosing the right cluster heads (CHs) and the best pathways in a WSN. The presented model includes a fitness value-based clustering scheme for efficient CH selection to achieve this. The Deep Neural Network (DNN) algorithm is then used to carry out the routing operation. The suggested approach technique calculates a fitness function (FF) that consists of three variables, including node degree, base station distance, and residual energy. This fitness function aids in the WSN’s best route selection. Simulations were run to verify the presented model’s superiority in terms of network lifespan and energy efficiency, and the results demonstrated the model’s outstanding performance.