Energy-efficient artificial fish swarm-based clustering protocol for enhancing network lifetime in underwater wireless sensor networks

Underwater wireless sensor networks (UWSNs) face significant challenges, such as limited energy resources, high propagation delays, and harsh underwater environments. Efficient clustering can help address these challenges by grouping nearby nodes to minimize network fragmentation and balance energy consumption. However, placing gateways near the sink node can result in increased communication overhead and higher energy consumption in regions with concentrated data flow. To address these issues, we propose an energy-efficient artificial fish swarm-based clustering cognitive intelligence protocol (EAFSCCIP). EAFSCCIP leverages the collective behavior of artificial fish within a Bees algorithm framework, using a combination of heuristic and metaheuristic approaches for optimal cluster-head (CH) selection in each round. The protocol focuses on reducing energy consumption and extending network lifetime by considering real-time energy levels and the proximity of nodes for CH selection. Simulations have been executed in NS3 to validate and compare the performance of the proposed algorithm with the existing clustering protocols. The results indicate that EAFSCCIP significantly enhances the packet delivery ratio (PDR) by an average of 5.33% over existing methods and improves network lifetime by 6.54% compared to traditional protocols. It also reduces energy consumption by 25.6% and decreases packet loss by 50.5%, while achieving 20.4% higher throughput at the initial stage. These improvements make EAFSCCIP a promising solution for applications like acoustic monitoring in UWSNs, providing a balance between energy efficiency and reliable data transmission.