Pilot-Free End-to-End Underwater Acoustic Communication System Based on Autoencoder

The long delay spreads and significant Doppler effects of underwater acoustic (UWA) channels make the design of the UWA communication system more challenging. In this paper, we propose a learning-based end-to-end framework for UWA communications, leveraging a double feature extraction network (DFEN) for data preprocessing. The DFEN consists of an attention-based module and a mixer-based module for channel feature extraction and data feature extraction, respectively. Considering the diverse nature of UWA channels, we propose a stack-network with a two-step training strategy to enhance generalization. By avoiding the use of pilot information, the proposed network can learn data mapping that is robust to UWA channels. Evaluation results show that our proposed algorithm outperforms the baselines by at least 2 dB under bit error rate (BER) 10
⁻²
on the simulation channel, and surpasses the compared neural network by at least 5 dB under BER 5 × 10
⁻²
on the experiment channels.