Mobility Patterns in Underwater Sensor Networks

Underwater Wireless Sensor Networks (UWSNs) operates in complex aquatic environments where energy constraints, unpredictable mobility, and acoustic communication delays pose significant challenges to efficient sensing and communication. This study addresses the problem of selecting appropriate node mobility models to enhance network performance - specifically energy consumption, coverage, and network lifetime. Three mobility strategies were investigated: Brownian random motion, Depth-Adjustable mobility, and Ocean Current-Based Drift. Each model was implemented in a simulated underwater environment using the Deep Deterministic Policy Gradient (DDPG) Reinforcement Learning(RL) algorithm. Simulations results indicate that the Depth-Adjustable Mobility Model consistently delivers the highest network lifetime. Its controlled vertical adjustments enable energy-efficient operation without significant loss in coverage. The Ocean Current-Based Drift Model offered a balanced performance, benefiting from adaptive current-following strategies. However, it lagged behind the depth-adjusted model in overall longevity. The Brownian Motion Model, proved to be the least efficient - exhibiting moderate energy consumption and reduced lifetime due to its inherently chaotic motion and lack of structured routing logic. The study finds that controlled mobility, especially depth-adjustable movement, is optimal for long-term and energy-sensitive underwater deployments. But models that simulate random or purely environmental drift are less effective for structured monitoring tasks.