关于水下语义无线通信的韧性
摘要
Underwater wireless communications 面临显著挑战 due to propagation constraints，限制了 traditional radio and optical technologies 的 effectiveness。Long-range acoustic communications 支持 distances up to a few kilometers，但 suffer from low bandwidth、high error ratios and multipath interference。Semantic communications focus on transmitting extracted semantic features rather than raw data，present a promising solution by significantly reducing the volume of data transmitted over the wireless link。本 paper evaluates SAGE，一种 semantic-oriented communications framework 的 resilience，该 framework combines semantic processing with Generative Artificial Intelligence (GenAI) to compress and transmit image data as textual descriptions over acoustic links。为 assess robustness，我们使用 custom-tailored simulator 引入 underwater acoustic channels 中 observed 的 character errors。Evaluation results show that SAGE 在 varying error conditions 下 能够 successfully reconstruct meaningful image content，highlighting its potential for robust and efficient underwater wireless communication in harsh environments。
引用
@article{arxiv.2506.23350,
title = {On the Resilience of Underwater Semantic Wireless Communications},
author = {João Pedro Loureiro and Patrícia Delgado and Tomás Feliciano Ribeiro and Filipe B. Teixeira and Rui Campos},
journal= {arXiv preprint arXiv:2506.23350},
year = {2025}
}