Abstract
The underwater wireless sensor network (UWSN) has been widely employed in scientific and commercial applications. The underwater wireless optical communication (UWOC) provides high transmission data rate, high security, and low latency communication to link the nodes in a UWSN. However, the link length of the UWOC systems is severely affected by large absorption and scattering losses that reduce the received optical signal-to-noise ratio (OSNR). The advanced signal processing technologies are hereby highly desired in UWOC systems to improve the electrical signal-to-noise (SNR) from the optical signal. In this paper, for the first time we have applied the Deep Echo State Network (DeepESN) in high-speed UWOC system to assist signal processing for both PAM and QAM-OFDM signals. We have experimentally verified the effectiveness of DeepESN in UWOC systems and demonstrated over 40.5 meters error-free underwater transmission with 100-167 Mbps data-rate.
| Original language | English |
|---|---|
| Title of host publication | 2022 Asia Communications and Photonics Conference (ACP) |
| Publisher | IEEE |
| Pages | 710-713 |
| Number of pages | 4 |
| ISBN (Electronic) | 9781665481557 |
| DOIs | |
| Publication status | Published - 10 Apr 2023 |
Keywords
- Deep echo state network
- Digital signal processing
- Machine learning
- Underwater wireless optical communication
ASJC Scopus subject areas
- Computer Networks and Communications
- Electrical and Electronic Engineering