Deep Learning Based High-Speed Underwater Acoustic Communication System Design
Access Status
Open access
Date
2025Supervisor
Yue Rong
Kit Yan Chan
Type
Thesis
Award
PhD
Metadata
Show full item recordFaculty
Science and Engineering
School
School of Electrical Engineering, Computing and Mathematical Sciences
Collection
Abstract
Underwater acoustic (UA) communication is vital for applications like oceanographic data collection and defence but faces challenges like limited bandwidth and multipath interference. This research enhances the conventional OFDM method with a neural network (NN)-based approach, leveraging CNNs, LSTMs, and MLPs for improved channel estimation and adaptability. Experiments in simulations, tank tests, and river trials demonstrated the superiority of the NN-based receiver, particularly under high non-linearity and Doppler effects, advancing UA communication reliability and adaptability.