Sliding window approach for continuous signal processing with nonlinear Fourier transform

Standard

Sliding window approach for continuous signal processing with nonlinear Fourier transform. / Chekhovskoy, I. S.; Sedov, E. V.; Fedoruk, M. P. et al.

In: Physica D: Nonlinear Phenomena, Vol. 497, 135388, 11.2026.

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

BibTeX

@article{59b90a587afd4963aff03eba5fb65943,

title = "Sliding window approach for continuous signal processing with nonlinear Fourier transform",

abstract = "We propose a novel approach to enhance the accuracy and efficiency of continuous signal processing using the nonlinear Fourier transform (NFT) with vanishing boundary conditions. Our approach utilizes a sliding window to segment the continuous signal, enabling subsequent processing of each segment via the NFT. To mitigate inter-segment interference, chromatic dispersion is compensated for within each window, which substantially enhances the NFT's performance while keeping the receiver-side processing computationally competitive. We describe the general concept and numerical methods underlying this approach, demonstrating its good performance in different configurations. We discuss specific results for 16-quadrature amplitude modulation (16-QAM) signal with single and dual polarizations transmitted over a 12-span standard single-mode fiber link with 80 km per span, with and without noise added by erbium-doped fiber amplifiers.",

keywords = "Nonlinear Fourier transform, Optical communication, Signal processing, Оптическая связь, Нелинейное преобразование Фурье, Обработка сигналов",

author = "Chekhovskoy, {I. S.} and Sedov, {E. V.} and Fedoruk, {M. P.} and Turitsyn, {S. K.}",

note = "I.S. Chekhovskoy, E.V. Sedov, M.P. Fedoruk, S.K. Turitsyn, Sliding window approach for continuous signal processing with nonlinear Fourier transform, Physica D: Nonlinear Phenomena, Volume 497, 2026, 135388, ISSN 0167-2789, https://doi.org/10.1016/j.physd.2026.135388. This work was supported by the RSF grant 25-61-00010, https://rscf.ru/project/25-61-00010/. S.K.T. acknowledges support by the Engineering and Physical Sciences Research Council grant EP/W002868/1.",

year = "2026",

month = nov,

doi = "10.1016/j.physd.2026.135388",

language = "English",

volume = "497",

journal = "Physica D: Nonlinear Phenomena",

issn = "0167-2789",

publisher = "Elsevier Science Publishing Company, Inc.",

}

RIS

TY - JOUR

T1 - Sliding window approach for continuous signal processing with nonlinear Fourier transform

AU - Chekhovskoy, I. S.

AU - Sedov, E. V.

AU - Fedoruk, M. P.

AU - Turitsyn, S. K.

N1 - I.S. Chekhovskoy, E.V. Sedov, M.P. Fedoruk, S.K. Turitsyn, Sliding window approach for continuous signal processing with nonlinear Fourier transform, Physica D: Nonlinear Phenomena, Volume 497, 2026, 135388, ISSN 0167-2789, https://doi.org/10.1016/j.physd.2026.135388. This work was supported by the RSF grant 25-61-00010, https://rscf.ru/project/25-61-00010/. S.K.T. acknowledges support by the Engineering and Physical Sciences Research Council grant EP/W002868/1.

PY - 2026/11

Y1 - 2026/11

N2 - We propose a novel approach to enhance the accuracy and efficiency of continuous signal processing using the nonlinear Fourier transform (NFT) with vanishing boundary conditions. Our approach utilizes a sliding window to segment the continuous signal, enabling subsequent processing of each segment via the NFT. To mitigate inter-segment interference, chromatic dispersion is compensated for within each window, which substantially enhances the NFT's performance while keeping the receiver-side processing computationally competitive. We describe the general concept and numerical methods underlying this approach, demonstrating its good performance in different configurations. We discuss specific results for 16-quadrature amplitude modulation (16-QAM) signal with single and dual polarizations transmitted over a 12-span standard single-mode fiber link with 80 km per span, with and without noise added by erbium-doped fiber amplifiers.

AB - We propose a novel approach to enhance the accuracy and efficiency of continuous signal processing using the nonlinear Fourier transform (NFT) with vanishing boundary conditions. Our approach utilizes a sliding window to segment the continuous signal, enabling subsequent processing of each segment via the NFT. To mitigate inter-segment interference, chromatic dispersion is compensated for within each window, which substantially enhances the NFT's performance while keeping the receiver-side processing computationally competitive. We describe the general concept and numerical methods underlying this approach, demonstrating its good performance in different configurations. We discuss specific results for 16-quadrature amplitude modulation (16-QAM) signal with single and dual polarizations transmitted over a 12-span standard single-mode fiber link with 80 km per span, with and without noise added by erbium-doped fiber amplifiers.

KW - Nonlinear Fourier transform

KW - Optical communication

KW - Signal processing

KW - Оптическая связь

KW - Нелинейное преобразование Фурье

KW - Обработка сигналов

UR - https://www.mendeley.com/catalogue/32d57e01-384e-3e19-9346-93b3882bfc55/

UR - https://www.scopus.com/pages/publications/105049317060

U2 - 10.1016/j.physd.2026.135388

DO - 10.1016/j.physd.2026.135388

M3 - Article

VL - 497

JO - Physica D: Nonlinear Phenomena

JF - Physica D: Nonlinear Phenomena

SN - 0167-2789

M1 - 135388

ER -

ID: 83162750