Machine Learning in Short-Reach Optical Systems - A Comprehensive Survey - Abstract

Aug 4, 2026

Upload by Aria

Extensive research investigates machine learning (ML) for direct-detected and (self)-coherent short-reach optical communication. Targeted tasks include bandwidth request prediction, signal-quality monitoring, fault detection, traffic prediction, and DSP-based equalization. ML can capture stochastic behaviors where deterministic methods fail, but gains from common DSP equalizers (FFEs/DFEs, Volterra nonlinear equalizers) are often small while complexity becomes excessive for cost-sensitive short-reach scenarios such as PONs. The survey emphasizes time-series models, temporal dependency capture, irregular/nonlinear handling, and variable intervals, proposing a taxonomy and future directions to reduce hardware complexity.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages22

Reading Time55 min

AccessFree to download

Introduction

Machine learning applications in short-reach optical communications

Time-series ML models and their advantages

Survey scope, taxonomy, and future research directions

hv photonics Review Machine Learning in Short-Reach Optical Systems: A Comprehensive Survey Chen Shao 1, *, Elias Giacoumidis 2, Syed Moktacim Billah 1, Shi Li 2, Jialei Li 2, Prashasti Sahu 3, André Richter 2, Michael Faerber 1 and Tobias Kaefer 1 Citation: Shao, C.; Giacoumidis, E.; Billah, S.M.; Li, S.; Li, J.; Sahu, P.; Richter, A.; Faerber, M.; Kaefer, T. Machine Learning in Short-Reach Optical Systems: A Comprehensive Survey. Photonics 2024, 11, 613 . [https://doi.org/10.3390/](https://doi.org/10.3390/)[ ](https://doi.org/10.3390/)photonics11070613 Received: 12 March 2024 Revised: 28 May 2024 Accepted: 29 May 2024 Published: 28 June 2024 Copyright: © 2024 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license ([https://](https://)[ ](https://)[creativecommons.org/licenses/by/](creativecommons.org/licenses/by/)[ ](creativecommons.org/licenses/by/)[4.0/](4.0/)) . 1 Department of Economics and Management, Karlsruhe Institute of Technology (KIT), 76131 Karlsruhe, Germany; [ge59wah@mytum.de](ge59wah@mytum.de) (S.M.B.); [michael.faerber@kit.edu](michael.faerber@kit.edu) (M.F.); [tobias.kaefer@kit.edu](tobias.kaefer@kit.edu) (T.K.) 2 VPIphotonics GmbH, Hallerstraße 6, 10587 Berlin, Germany; [elias.giacoumidis@vpiphotonics.com](elias.giacoumidis@vpiphotonics.com) (E.G.); [shi.li@vpiphotonics.com](shi.li@vpiphotonics.com) (S.L.); [jia.li9@kit.edu](jia.li9@kit.edu) (J.L.); [andre.richter@vpiphotonics.com](andre.richter@vpiphotonics.com) (A.R.) 3 Electronic and Information Engineering, Technical University of Chemnitz, Str. der Nationen 62, 09111 Chemnitz, Germany; [prashastisahu13@gmail.com](prashastisahu13@gmail.com) * Correspondence: [chen.shao2@kit.edu](chen.shao2@kit.edu) Abstract: Recently, extensive research has been conducted to explore the utilization of machine learning (ML) algorithms in various direct-detected and (self)-coherent short-reach communication applications. These applications encompass a wide range of tasks, including bandwidth request prediction, signal quality monitoring, fault detection, traffic prediction, and digital signal processing (DSP)-based equalization. As a versatile approach, ML demonstrates the ability to address stochastic phenomena in optical systems networks where deterministic methods may fall short. However, when it comes to DSP equalization algorithms such as feed-forward/decision-feedback equalizers (FFEs/DFEs) and Volterra-based nonlinear equalizers, their performance improvements are often marginal, and their complexity is prohibitively high, especially in cost-sensitive short-reach communications scenarios such as passive optical networks (PONs) . Time-series ML models offer distinct advantages over frequency-domain models in specific contexts. They excel in capturing temporal dependencies, handling irregular or nonlinear patterns effectively, and accommodating variable time intervals. Within this survey, we outline the application of ML techniques in short-reach communications, specifically emphasizing their utilization in high-bandwidth demanding PONs. We introduce a novel taxonomy for time-series methods employed in ML signal processing, providing a structured classification framework. Our taxonomy categorizes current time-series methods into four distinct groups: traditional methods, Fourier convolution-based methods, transformer-based models, and time-series convolutional networks. Finally, we highlight prospective research directions within this rapidly evolving field and outline specific solutions to mitigate the complexity associated with hardware implementations. We aim to pave the way for more practical and efficient deployment of ML approaches in short-reach optical communication systems by addressing complexity concerns. Keywords: machine learning; optical communications; passive optical network; equalization; op

FAQs

哪些短距离光通信场景会使用机器学习方法？

文中关注直检以及（自）相干的短距离通信，并覆盖带宽请求预测、信号质量监测、故障检测、流量预测以及DSP相关均衡等任务。

传统DSP均衡器使用ML后性能提升为什么常常有限？

文中指出，以FFEs/DFEs和基于Volterra的非线性均衡器为代表的方法，性能改进往往边际，但复杂度却可能很高，尤其在成本敏感的短距离通信（如PON）中更明显。

时间序列ML模型相比频域模型有哪些优势？

时间序列模型能更好地捕获时间依赖，能有效处理不规则或非线性模式，并能适应可变的时间间隔。

该综述如何组织与分类时间序列方法？

综述提出用于ML信号处理的时间序列方法新型分类体系，将现有时间序列方法划分为四类：传统方法、傅里叶卷积方法、基于Transformer的模型以及时间序列卷积网络。

Aug 4, 2026

Upload by Aria