Low computationally complex recurrent neural network for high speed optical fiber transmission

The demand for high speed data transmission has increased rapidly over the past few years, leading to the development of the data center concept. Considering that vertical cavity surface emitting lasers (VCSELs) based optical interconnect is evolving to 100 Gb/s, relative intensity noise and mode pa...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Optics communications Ročník 441; s. 121 - 126
Hlavní autoři: Zhou, Qingyi, Yang, Chuanchuan, Liang, Anzhong, Zheng, Xiaolong, Chen, Zhangyuan
Médium: Journal Article
Jazyk:angličtina
Vydáno: Elsevier B.V 15.06.2019
Témata:
Low-computational-complexity
Nonlinear equalizer
Optical interconnect
PAM-4
Recurrent neural network
Vertical cavity surface emitting lasers (VCSELs)
20-060
Recurrent neural network
Low-computational-complexity
Optical interconnect
PAM-4
20-100
Vertical cavity surface emitting lasers (VCSELs)
Nonlinear equalizer
ISSN:0030-4018, 1873-0310
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:The demand for high speed data transmission has increased rapidly over the past few years, leading to the development of the data center concept. Considering that vertical cavity surface emitting lasers (VCSELs) based optical interconnect is evolving to 100 Gb/s, relative intensity noise and mode partition noise are becoming significant, which cannot be equalized by conventional equalizers efficiently. On the other hand, optical fiber suffers from inter-symbol interference (ISI) and nonlinear channel response, making the equalization process even more challenging. Recently, several machine learning techniques have already been applied to recover signal from nonlinear distortions. However, previous experiments mainly focused on achieving low BER and have neglected the computational complexity. In this paper, we have designed a recurrent neural network (RNN) based equalizer. The equalizer is tested over a PAM-4 modulated VCSEL-MMF optical interconnect link, and shows BER performance improvement over equalizers based on ANN. A variant, Half-RNN, is also proposed and tested, whose computational complexity is 70% lower than ANN with similar BER performance. Our experiments provide guidance for designing neural network structure when using deep learning for equalization, justifying the significance of our work.
ISSN:0030-4018
1873-0310
DOI:10.1016/j.optcom.2019.02.037