Enabling High-Throughput and Energy-Efficient Underwater Networks through Acoustic-Magnetic Multimodal MIMO Communications

Underwater networks often rely on acoustic communication to overcome high electromagnetic loss and achieve long-range connectivity. However, the high delay of acoustic signals can lead to low throughput, and traditional routine protocols may not effectively balance network performance with energy consumption. This paper proposes an underwater network that employs acoustic-magnetic multimodal MIMO communication to achieve high throughput and energy efficiency. Rather than relying solely on acoustic approaches, we introduce two modules: magnetic induction (MI) communication for in-cluster node synchronization, and acoustic signals for cooperative MIMO between clusters and the base station node. This design significantly enhances network throughput. Additionally, we propose a Q-learning-based protocol to balance energy consumption with performance in this network. Simulation results show nearly ten times of the network link throughput over existing acoustic schemes, along with significant improvements in network connectivity and energy efficiency when using the proposed Q-learning routing protocol.