Use of complex fully connected neural networks to compensate for nonlinear effects in fibre-optic communication lines

S. A. Bogdanov; O. S. Sidelnikov

doi:10.1070/QEL17518

Use of complex fully connected neural networks to compensate for nonlinear effects in fibre-optic communication lines

Research output: Contribution to journal › Article › peer-review

Abstract

A scheme is proposed for processing optical signals in a receiver of a communication system, based on complex fully connected neural networks. The influence of the main characteristics of the neural network on the efficiency of nonlinear distortion compensation is studied. A significant advantage of the proposed scheme over real-valued neural networks is demonstrated.

Original language	English
Pages (from-to)	459-462
Number of pages	4
Journal	Quantum Electronics
Volume	51
Issue number	5
DOIs	https://doi.org/10.1070/QEL17518
Publication status	Published - May 2021

OECD FOS+WOS

2.02 ELECTRICAL ENG, ELECTRONIC ENG
2.05 MATERIALS ENGINEERING
1.03 PHYSICAL SCIENCES AND ASTRONOMY
1.03.UB PHYSICS, APPLIED
2.02.IQ ENGINEERING, ELECTRICAL & ELECTRONIC

Access to Document

10.1070/QEL17518

Cite this

@article{64680ac7fd944dc39f9b3a29153ced33,

title = "Use of complex fully connected neural networks to compensate for nonlinear effects in fibre-optic communication lines",

abstract = "A scheme is proposed for processing optical signals in a receiver of a communication system, based on complex fully connected neural networks. The influence of the main characteristics of the neural network on the efficiency of nonlinear distortion compensation is studied. A significant advantage of the proposed scheme over real-valued neural networks is demonstrated. ",

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

year = "2021",

month = may,

doi = "10.1070/QEL17518",

language = "English",

volume = "51",

pages = "459--462",

journal = "Quantum Electronics",

issn = "1063-7818",

publisher = "Turpion Ltd.",

number = "5",

}

TY - JOUR

T1 - Use of complex fully connected neural networks to compensate for nonlinear effects in fibre-optic communication lines

AU - Bogdanov, S. A.

AU - Sidelnikov, O. S.

PY - 2021/5

Y1 - 2021/5

N2 - A scheme is proposed for processing optical signals in a receiver of a communication system, based on complex fully connected neural networks. The influence of the main characteristics of the neural network on the efficiency of nonlinear distortion compensation is studied. A significant advantage of the proposed scheme over real-valued neural networks is demonstrated.

AB - A scheme is proposed for processing optical signals in a receiver of a communication system, based on complex fully connected neural networks. The influence of the main characteristics of the neural network on the efficiency of nonlinear distortion compensation is studied. A significant advantage of the proposed scheme over real-valued neural networks is demonstrated.

UR - http://www.scopus.com/inward/record.url?scp=85106225572&partnerID=8YFLogxK

U2 - 10.1070/QEL17518

DO - 10.1070/QEL17518

M3 - Article

AN - SCOPUS:85106225572

VL - 51

SP - 459

EP - 462

JO - Quantum Electronics

JF - Quantum Electronics

SN - 1063-7818

IS - 5

ER -

Use of complex fully connected neural networks to compensate for nonlinear effects in fibre-optic communication lines

Abstract

OECD FOS+WOS

Access to Document

Other files and links

Fingerprint

Cite this