Standard

ML-Assisted Subband Learned Digital Backpropagation for Nonlinearity Compensation in Wideband Optical Systems. / Shevelev, Evgeny; Sidelnikov, Oleg; Danilko, Vitaly et al.

In: Journal of Lightwave Technology, Vol. 44, No. 18, 04.08.2026, p. 8023-8035.

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

Harvard

APA

Vancouver

Shevelev E, Sidelnikov O, Danilko V, Fedoruk M, Redyuk A. ML-Assisted Subband Learned Digital Backpropagation for Nonlinearity Compensation in Wideband Optical Systems. Journal of Lightwave Technology. 2026 Aug 4;44(18):8023-8035. doi: 10.1109/JLT.2026.3720192

Author

Shevelev, Evgeny ; Sidelnikov, Oleg ; Danilko, Vitaly et al. / ML-Assisted Subband Learned Digital Backpropagation for Nonlinearity Compensation in Wideband Optical Systems. In: Journal of Lightwave Technology. 2026 ; Vol. 44, No. 18. pp. 8023-8035.

BibTeX

@article{b107392f5b264dd29f508dc1e28f09a5,
title = "ML-Assisted Subband Learned Digital Backpropagation for Nonlinearity Compensation in Wideband Optical Systems",
abstract = "Digital backpropagation (DBP) is one of the most effective techniques for compensating nonlinear distortions in coherent optical fiber communication systems. However, its practical application to wideband transmission remains limited by high computational complexity caused by large channel memory and the requirement for fine spatial discretization. In this work, we propose a subband-based learned digital backpropagation (SbL-DBP) framework for wideband optical transmission systems. The received signal is decomposed into multiple subbands, enabling independent frequency-domain compensation of the chromatic dispersion with reduced effective channel memory and lower computational complexity. Nonlinear intra- and inter-subband interactions are addressed in the time domain using a trainable multi-input multi-output filtering structure. The parameters of the proposed framework are jointly optimized using end-to-end gradient-based learning. In addition, sparsification techniques are employed to remove insignificant coefficients and further reduce computational complexity. Numerical simulations of an 11×40 Gbaud WDM RRC-16QAM 20×100 km transmission system demonstrate that the proposed method provides a superior performance-complexity trade-off compared to conventional DBP and enhanced DBP. In the low- and medium-complexity regimes, SbL-DBP provides higher signal-to-noise ratio gains while requiring fewer propagation steps.",
keywords = "Digital backpropagation, digital signal processing, fiber-optic communication links, machine learning, nonlinear distortion compensation, nonlinearity, subband processing, Цифровое обратное распространение, цифровая обработка сигналов, волоконно-оптические линии связи, машинное обучение, нелинейность, компенсация нелинейных искажений, поддиапазонная обработка",
author = "Evgeny Shevelev and Oleg Sidelnikov and Vitaly Danilko and Mikhail Fedoruk and Alexey Redyuk",
note = "E. Shevelev, O. Sidelnikov, V. Danilko, M. Fedoruk and A. Redyuk, {"}ML-Assisted Subband Learned Digital Backpropagation for Nonlinearity Compensation in Wideband Optical Systems,{"} in Journal of Lightwave Technology, vol. 44, no. 18, pp. 8023-8035, 15 Sept.15, 2026, doi: 10.1109/JLT.2026.3720192. This work was supported by Russian Science Foundation under Grant 25-61-00010.",
year = "2026",
month = aug,
day = "4",
doi = "10.1109/JLT.2026.3720192",
language = "English",
volume = "44",
pages = "8023--8035",
journal = "Journal of Lightwave Technology",
issn = "0733-8724",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "18",

}

RIS

TY - JOUR

T1 - ML-Assisted Subband Learned Digital Backpropagation for Nonlinearity Compensation in Wideband Optical Systems

AU - Shevelev, Evgeny

AU - Sidelnikov, Oleg

AU - Danilko, Vitaly

AU - Fedoruk, Mikhail

AU - Redyuk, Alexey

N1 - E. Shevelev, O. Sidelnikov, V. Danilko, M. Fedoruk and A. Redyuk, "ML-Assisted Subband Learned Digital Backpropagation for Nonlinearity Compensation in Wideband Optical Systems," in Journal of Lightwave Technology, vol. 44, no. 18, pp. 8023-8035, 15 Sept.15, 2026, doi: 10.1109/JLT.2026.3720192. This work was supported by Russian Science Foundation under Grant 25-61-00010.

PY - 2026/8/4

Y1 - 2026/8/4

N2 - Digital backpropagation (DBP) is one of the most effective techniques for compensating nonlinear distortions in coherent optical fiber communication systems. However, its practical application to wideband transmission remains limited by high computational complexity caused by large channel memory and the requirement for fine spatial discretization. In this work, we propose a subband-based learned digital backpropagation (SbL-DBP) framework for wideband optical transmission systems. The received signal is decomposed into multiple subbands, enabling independent frequency-domain compensation of the chromatic dispersion with reduced effective channel memory and lower computational complexity. Nonlinear intra- and inter-subband interactions are addressed in the time domain using a trainable multi-input multi-output filtering structure. The parameters of the proposed framework are jointly optimized using end-to-end gradient-based learning. In addition, sparsification techniques are employed to remove insignificant coefficients and further reduce computational complexity. Numerical simulations of an 11×40 Gbaud WDM RRC-16QAM 20×100 km transmission system demonstrate that the proposed method provides a superior performance-complexity trade-off compared to conventional DBP and enhanced DBP. In the low- and medium-complexity regimes, SbL-DBP provides higher signal-to-noise ratio gains while requiring fewer propagation steps.

AB - Digital backpropagation (DBP) is one of the most effective techniques for compensating nonlinear distortions in coherent optical fiber communication systems. However, its practical application to wideband transmission remains limited by high computational complexity caused by large channel memory and the requirement for fine spatial discretization. In this work, we propose a subband-based learned digital backpropagation (SbL-DBP) framework for wideband optical transmission systems. The received signal is decomposed into multiple subbands, enabling independent frequency-domain compensation of the chromatic dispersion with reduced effective channel memory and lower computational complexity. Nonlinear intra- and inter-subband interactions are addressed in the time domain using a trainable multi-input multi-output filtering structure. The parameters of the proposed framework are jointly optimized using end-to-end gradient-based learning. In addition, sparsification techniques are employed to remove insignificant coefficients and further reduce computational complexity. Numerical simulations of an 11×40 Gbaud WDM RRC-16QAM 20×100 km transmission system demonstrate that the proposed method provides a superior performance-complexity trade-off compared to conventional DBP and enhanced DBP. In the low- and medium-complexity regimes, SbL-DBP provides higher signal-to-noise ratio gains while requiring fewer propagation steps.

KW - Digital backpropagation

KW - digital signal processing

KW - fiber-optic communication links

KW - machine learning

KW - nonlinear distortion compensation

KW - nonlinearity

KW - subband processing

KW - Цифровое обратное распространение

KW - цифровая обработка сигналов

KW - волоконно-оптические линии связи

KW - машинное обучение

KW - нелинейность

KW - компенсация нелинейных искажений

KW - поддиапазонная обработка

UR - https://www.mendeley.com/catalogue/c9282235-a45f-3520-a7b0-09aacb2d58fe/

UR - https://www.scopus.com/pages/publications/105046838278

U2 - 10.1109/JLT.2026.3720192

DO - 10.1109/JLT.2026.3720192

M3 - Article

VL - 44

SP - 8023

EP - 8035

JO - Journal of Lightwave Technology

JF - Journal of Lightwave Technology

SN - 0733-8724

IS - 18

ER -

ID: 85312231