A Channel Adaptive MIMO Spatial Multiplexing UWOC System Based on Deep Learning Encoder/Decoder

The absorption, scattering and turbulence have a significant and time-varying impact on light transmission underwater. The light-emitting diodes (LED) as a light source in an underwater wireless optical communication (UWOC) system significantly minimize the requirement for the transceiver alignment,...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:2023 4th Information Communication Technologies Conference (ICTC) s. 27 - 33
Hlavní autoři: Zeng, Lingbiao, Yin, Hongxi, Wang, Xinge, Ji, Xiuyang
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: IEEE 17.05.2023
Témata:
Absorption
Adaptive systems
channel adaptive
Deep learning
Light emitting diodes
MIMO
Optical fiber communication
Scattering
spatial multiplexing
underwater wireless optical communication
Wireless communication
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:The absorption, scattering and turbulence have a significant and time-varying impact on light transmission underwater. The light-emitting diodes (LED) as a light source in an underwater wireless optical communication (UWOC) system significantly minimize the requirement for the transceiver alignment, but the capacity and the reliability of the system are constrained by the narrow bandwidth of the LED. To this end, this paper proposes a channel adaptive multiple-input multiple-output (MIMO) spatial multiplexing (SM) system based on the deep learning encoder/decoder to adapt to the channel variation and optimize system performance by learning different water channels and different modulation and demodulation methods. Our simulation experiment demonstrates that EDCC has lower SER performance and better robustness in turbulent channels than conventional LS-MMSE, LS-ML, and deep learning based DNN-D and DNN-ED algorithms.
DOI:10.1109/ICTC57116.2023.10154825