A Machine Learning-Based Cluster Head Selection Method for Data Transmission and Energy Efficiency in Underwater Wireless Sensor Networks

Resumen

Underwater wireless sensor networks (WSNs) have become more important due to their various applications. To maintain network connectivity and transmission efficiency, routing protocols are crucial for UWSN. Efficient energy consumption and reliable data transmission are critical challenges in Underwater Wireless Sensor Networks (UWSNs) due to harsh underwater environment and limited energy resources. In an existing data transmission method, a routing protocol called multi-hop is used, which leads to high energy consumption. In the proposed method, all the sensor nodes are organized into different clusters, a random sensor node is selected as a cluster head who gathers and sends data among the cluster. Selection of cluster-head can be done by using Machine Learning based Cluster Head Selection Method (ML-CHSM). This cluster-head selection method reduces energy usage and increases durability of the network. The machine learning-based cluster head selection method aims to optimize energy consumption, and improve network's operational lifetime, data transmission reliability. In this paper, the usage of machine learning techniques is proposed to increase energy level and optimize transmission of data in UWSNs. Through simulations, the effectiveness of overall efficiency and data transmission reliability is shown in UWSNs.