Energy-Efficient Routing in Underwater Wireless Sensor Networks Using Swarm-Intelligent Cluster Optimization

In recent years, Underwater Wireless Sensor Networks (UWSNs) have been increasingly implemented in applications, such as smart cities, environmental monitoring, and underwater routing. Existing clustering-based routing models, including Low-Energy Adaptive Clustering Hierarchy (LEACH) and Energy Optimization Routing using improved Artificial Bee Colony (EOR-iABC), often suffer from constrained global search, random Cluster Head (CH) selection, and a lack of adaptability to changing node conditions. To overcome these limitations, this research proposes an EnergyEfficient Routing with Swarm Intelligent Cluster Optimization (EER-SICO) model for hierarchical UWSNs. In EER-SICO, the SICO integrates Sine Cosine Algorithm (SCA) for the global exploration of potential CHs and utilizes Multi-Agent Reinforcement Learning (MARL) for an adaptive CH rotation depending on learned node behaviors and network dynamics. Additionally, a Q-learning-based multi-hop routing strategy enhances energy-aware data forwarding, and a Markov-based failure prediction mechanism ensures continuity in the CH functionality. Experimental results show that EER-SICO significantly enhances energy efficiency (17.9%) and reduces end-to-end delay (0.98%) over existing EOR-iABC respectively.