Application of complex fully connected neural networks to compensate for nonlinearity in fibre-optic communication lines with polarisation division multiplexing

Standard

Application of complex fully connected neural networks to compensate for nonlinearity in fibre-optic communication lines with polarisation division multiplexing. / Bogdanov, S. A.; Sidelnikov, O. S.; Redyuk, A. A.

в: Quantum Electronics, Том 51, № 12, 5, 12.2021, стр. 1076-1080.

Результаты исследований: Научные публикации в периодических изданиях › статья › Рецензирование

BibTeX

@article{8bffab5b1adc4032a368ec06aa44a371,

title = "Application of complex fully connected neural networks to compensate for nonlinearity in fibre-optic communication lines with polarisation division multiplexing",

abstract = "A scheme is proposed to compensate for nonlinear distortions in extended fibre-optic communication lines with polarisation division multiplexing, based on fully connected neural networks with complex-valued arithmetic. The activation function of the developed scheme makes it possible to take into account the nonlinear interaction of signals from different polarisation components. This scheme is compared with a linear one and a neural network that processes signals of different polarisations independently, and the superiority of the proposed neural network architecture is demonstrated.",

author = "Bogdanov, {S. A.} and Sidelnikov, {O. S.} and Redyuk, {A. A.}",

note = "Acknowledgements. The work was supported by the RF President{\textquoteright}s Grants Council (State Support to Young Russian Scientists Programme, Grant No. MK-915.2020.9). The work of S.A. Bogdanov was supported by the state assignment for fundamental research (FSUS-2020-0034). The work of A.A. Redyuk was supported by the Ministry of Science and Higher Education of the Russian Federation (Project No. FSUS-2021-0015). Publisher Copyright: {\textcopyright} 2021 Kvantovaya Elektronika and IOP Publishing Limited.",

year = "2021",

month = dec,

doi = "10.1070/QEL17656",

language = "English",

volume = "51",

pages = "1076--1080",

journal = "Quantum Electronics",

issn = "1063-7818",

publisher = "Turpion Ltd.",

number = "12",

}

RIS

TY - JOUR

T1 - Application of complex fully connected neural networks to compensate for nonlinearity in fibre-optic communication lines with polarisation division multiplexing

AU - Bogdanov, S. A.

AU - Sidelnikov, O. S.

AU - Redyuk, A. A.

N1 - Acknowledgements. The work was supported by the RF President’s Grants Council (State Support to Young Russian Scientists Programme, Grant No. MK-915.2020.9). The work of S.A. Bogdanov was supported by the state assignment for fundamental research (FSUS-2020-0034). The work of A.A. Redyuk was supported by the Ministry of Science and Higher Education of the Russian Federation (Project No. FSUS-2021-0015). Publisher Copyright: © 2021 Kvantovaya Elektronika and IOP Publishing Limited.

PY - 2021/12

Y1 - 2021/12

N2 - A scheme is proposed to compensate for nonlinear distortions in extended fibre-optic communication lines with polarisation division multiplexing, based on fully connected neural networks with complex-valued arithmetic. The activation function of the developed scheme makes it possible to take into account the nonlinear interaction of signals from different polarisation components. This scheme is compared with a linear one and a neural network that processes signals of different polarisations independently, and the superiority of the proposed neural network architecture is demonstrated.

AB - A scheme is proposed to compensate for nonlinear distortions in extended fibre-optic communication lines with polarisation division multiplexing, based on fully connected neural networks with complex-valued arithmetic. The activation function of the developed scheme makes it possible to take into account the nonlinear interaction of signals from different polarisation components. This scheme is compared with a linear one and a neural network that processes signals of different polarisations independently, and the superiority of the proposed neural network architecture is demonstrated.

UR - http://www.scopus.com/inward/record.url?scp=85122510545&partnerID=8YFLogxK

UR - https://www.elibrary.ru/item.asp?id=47902491

UR - https://www.mendeley.com/catalogue/36573f8f-07df-3b62-83b4-193290f2581f/

U2 - 10.1070/QEL17656

DO - 10.1070/QEL17656

M3 - Article

AN - SCOPUS:85122510545

VL - 51

SP - 1076

EP - 1080

JO - Quantum Electronics

JF - Quantum Electronics

SN - 1063-7818

IS - 12

M1 - 5

ER -

ID: 35198644