An Energy Efficient Multi-Level Clustering and Optimized Path Planning With Data Fusion (EEMLC-OPPDF) for Autonomous Underwater Vehicles (AUVs) in Underwater Wireless Sensor Networks

Abstract

The major network elements of underwater wireless sensor networks (UWSNs) are their energy-limited sensor nodes, which are used to monitor harsh underwater environments. Ensuring UWSN sustainability and functionality necessitates reducing the energy constraints of these nodes. Continuous data collection and transmission drastically reduce the energy sources of nodes. To address this, autonomous underwater vehicles (AUVs) are used in UWSNs to collect data from nodes, thereby reducing transmission costs and solving energy constraints. This research introduces an energy-efficient multi-level clustering and cluster head (CH) selection procedure, combined with an optimized path planning strategy for AUVs using the Glowworm Swarm Optimization (GSO) algorithm. Additionally, a data fusion model based on an Enhanced Back Propagation Neural Network (EBPNN) is proposed to eliminate redundant data, thus reducing the transmission load. The novel EEMLC-OPPDF model is experimentally validated, demonstrating superior performance in terms of energy consumption and network lifetime compared to state-of-the-art techniques. This work contributes a comprehensive solution to optimize energy usage and extend the operational life of UWSNs.