Universal Modem Generation with Inherent Adaptability to Variant Underwater Acoustic Channels: a Data-Driven Perspective

TL;DR

This paper introduces UWAModNet, a data-driven neural network-based modem design that adapts to underwater acoustic channels, outperforming traditional OFDM in handling Doppler effects and improving communication reliability.

Contribution

The paper presents a novel multi-resolution CNN for adaptive modem design in underwater acoustic channels, with a new optimization criterion and training strategy for enhanced performance.

Findings

Outperforms zero-padded OFDM in sub-channel rate

Achieves lower bit error rate under severe Doppler effects

Demonstrates adaptability to varying underwater channel conditions

Abstract

In underwater acoustic (UWA) communication, orthogonal frequency division multiplexing (OFDM) is commonly employed to mitigate the inter-symbol interference (ISI) caused by delay spread. However, path-specific Doppler effects in UWA channels could result in significant inter-carrier interference (ICI) in the OFDM system. To address this problem, we introduce a multi-resolution convolutional neural network (CNN) named UWAModNet in this paper, designed to optimize the modem structure, specifically modulation and demodulation matrices. Based on a trade-off between the minimum and the average equivalent sub-channel rate, we propose an optimization criterion suitable to evaluate the performance of our learned modem. Additionally, a two-stage training strategy is developed to achieve quasi-optimal results. Simulations indicate that the learned modem outperforms zero-padded OFDM (ZP-OFDM) in terms of equivalent sub-channel rate and bit error rate, even under more severe Doppler effects during testing compared to training.