An Enhanced Lifespan of WSN Using Hybrid Fuzzy-Machine Learning-Based Clustering Process

Wireless sensor networks (WSN) are networks formed by the sensing nodes, capable of observing, recording, and communicating the intended information to the base station. The network of sensor nodes forms a cluster based on multiple characteristics, namely, the sensor's nature, a measuring factor, the distance between neighboring nodes, geographical linkage, etc. Various approaches have been employed to perform the clustering process and improve the lifetime of the sensor nodes. The sensor nodes communicate through the Cluster Head (CH) to mitigate energy consumption due to the route discovery process. The successful transmission of data and the clustering process drain the sensor nodes, reducing network lifetime. This research manuscript introduces a novel clustering process that mitigates overhead by considering residual energy during the clustering process. The clustering process uses fuzzy-based inference, which determines the residual energy and estimates the minimal energy for data communication between the source node and base station. The sensor nodes employ a machine learning process to categorize the data based on similarity. The categorized data is then communicated to the cluster head to mitigate the multiple data transmissions from the source node to the base station. The analysis of the proposed method employs fuzzy logic in determining the residual energy, and the machine learning algorithm in categorizing the data enhances the lifetime of the sensor nodes.