Session: 06-11-02 Ocean Engineering Technology - II

Submission Number: 180204

Experimental Study on OSDM Underwater Acoustic Communication Using Sparse Bayesian Channel Estimation 

To sustainably utilize the vast oceans, the use of underwater robotics, such as underwater unmanned vehicles (UUVs) or remotely operated vehicles (ROVs) for underwater infrastructure inspection, seafloor exploration, and offshore industry has been extensively studied. Underwater acoustic (UWA) communication is a promising technology to realize reliable wireless data transmission, which is necessary for the remote operation of UUVs or ROVs, since radio and optical waves cannot travel long distances in the underwater environment. However, the UWA channel is known as one of the most challenging communication environments due to its severe doubly selectivity compared with terrestrial radio communication.

 To cope with this severe channel, a number of modulation schemes, such as single carrier modulation (SC) and orthogonal frequency division multiplexing (OFDM), have been studied. As an alternative, we have proposed the use of orthogonal signal division multiplexing (OSDM). While SC and OFDM arrange the transmitted symbols in the time and frequency domains, respectively, OSDM is characterized by its periodic signal structure in the time and frequency domains, which provides robustness against the doubly selective channel. Previous studies revealed that exploiting the channel sparsity, which is the inherent nature of the UWA channel, improves demodulation performance. Since the existing sparse channel estimation methods adaptively operate on the received signal from the point of view of a deterministic one, they require prior knowledge of the communication environment or large computational complexity to fit some channel parameters. To deal with these problems, this study treats the unknown sparse channel as stochastic variables and enables the automatic determination of parameters to be set. Specifically, we employed sparse Bayesian learning, a framework for estimating sparse signals through evaluating a posterior distribution on the unknown variables, including channel parameters.
 In addition, we verified the proposed Bayesian channel estimation method in an actual sea trial by comparing it with the existing sparse channel estimation methods. Our experimental results demonstrated that our proposed method achieved a comparable communication performance with traditional sparse channel estimation methods, without any prior knowledge. We conclude that our proposed method becomes a viable sparse channel estimation in underwater acoustic communication.

Presenting Author: Ryoichi Ishijima University of Tsukuba

Presenting Author Biography: RYOICHI ISHIJIMA was born in Hiroshima, Japan, in 1999. He received the B.Eng. and M.Eng. degrees from the University of Tsukuba, Tsukuba, Japan, in 2023 and 2025, respectively, where he is currently pursuing the Ph.D. degree. His research interests include underwater acoustic communication. He received an Academic Encouragement Award from the Institute of Electrical Engineers of Japan in 2023 and was awarded a Research Fellowships for Young Scientists (DC1) from Japan Society for the Promotion of Science (JSPS) in 2025.

Authors:

Ryoichi Ishijima University of Tsukuba
Tadashi Ebihara University of Tsukuba
Naoto Wakatsuki University of Tsukuba