A Variational Auto-Encoder Model for Underwater Acoustic Channels
A Variational Auto-Encoder Model for Underwater Acoustic Channels
复制标题
DOI：
10.1145/3491315.3491330
复制
发表时间：
2021-11
期刊：
影响因子：
--
通讯作者：
Li Wei;Zhaohui Wang
中科院分区：
文献类型：
--
作者：
Li Wei;Zhaohui Wang
人们广泛需要一种具有高有效性和可重用性的水声（UWA）信道模型。在本文中，我们提出了一种基于变分自动编码器（VAE）的深度生成模型，该模型学习 UWA 通道脉冲响应（CIR）的抽象表示，并可以生成具有相似特征的 CIR 样本。提出了定制的训练过程，以避免模型崩溃并陷入梯度坑。使用现场实验数据集验证了所提出的深度生成模型。
An underwater acoustic (UWA) channel model with high validity and re-usability is widely demanded. In this paper, we propose a variational auto-encoder (VAE)-based deep generative model which learns an abstract representation of the UWA channel impulse responses (CIRs) and can generate CIR samples with similar features. A customized training process is proposed to avoid the model collapse and being trapped in a gradient pit. The proposed deep generative model is validated using field experimental data sets.