Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

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

Chen Shao, Elias Giacoumidis, Syed Moktacim Billah, Shi Li, Jialei Li, Prashasti Sahu, Andre Richter, Tobias Kaefer, Michael Faerber

TL;DR

The paper addresses how machine learning can improve short-reach optical communications, especially PONs, by surveying a broad range of ML-based DSP and prediction tasks. It introduces a structured taxonomy of time-series methods—traditional, Fourier-based, transformer-based, and time-series CNNs—and evaluates their applicability to bandwidth prediction, resource allocation, equalization, and fault detection under tight hardware constraints. A key contribution is a practical discussion of model compression techniques (distillation, vector quantization, pruning, quantization) to enable single-GPU deployment, along with insights into hardware-friendly architectures. Overall, the work maps the state of ML in short-reach optics, highlights complexity-hardware trade-offs, and points to concrete research directions for efficient, scalable deployment in cost-sensitive PON environments.

Abstract

In recent years, extensive research has been conducted to explore the utilization of machine learning 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, machine learning 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, 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). They excel in capturing temporal dependencies, handling irregular or nonlinear patterns effectively, and accommodating variable time intervals. Within this extensive survey, we outline the application of machine learning techniques in short-reach communications, specifically emphasizing their utilization in high-bandwidth demanding PONs. Notably, we introduce a novel taxonomy for time-series methods employed in machine learning 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 machine learning approaches in short-reach optical communication systems by addressing complexity concerns.

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

TL;DR

The paper addresses how machine learning can improve short-reach optical communications, especially PONs, by surveying a broad range of ML-based DSP and prediction tasks. It introduces a structured taxonomy of time-series methods—traditional, Fourier-based, transformer-based, and time-series CNNs—and evaluates their applicability to bandwidth prediction, resource allocation, equalization, and fault detection under tight hardware constraints. A key contribution is a practical discussion of model compression techniques (distillation, vector quantization, pruning, quantization) to enable single-GPU deployment, along with insights into hardware-friendly architectures. Overall, the work maps the state of ML in short-reach optics, highlights complexity-hardware trade-offs, and points to concrete research directions for efficient, scalable deployment in cost-sensitive PON environments.

Abstract

In recent years, extensive research has been conducted to explore the utilization of machine learning 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, machine learning 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, 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). They excel in capturing temporal dependencies, handling irregular or nonlinear patterns effectively, and accommodating variable time intervals. Within this extensive survey, we outline the application of machine learning techniques in short-reach communications, specifically emphasizing their utilization in high-bandwidth demanding PONs. Notably, we introduce a novel taxonomy for time-series methods employed in machine learning 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 machine learning approaches in short-reach optical communication systems by addressing complexity concerns.
Paper Structure (11 sections, 15 equations, 1 figure, 3 tables)

This paper contains 11 sections, 15 equations, 1 figure, 3 tables.

Table of Contents

  1. Introduction
  2. Applications in Short-Reach Systems
  3. DSP for Signal Equalization in Communication Systems
  4. Traditional Sequential ML Methods
  5. Advanced Sequential ML Methods
  6. Distortion Model:
  7. Temporal Convolution Neural Network
  8. Transformer-based Network
  9. Fourier Convolution Neural Network
  10. Model Compression
  11. Conclusion

Figures (1)

  • Figure S1: The overview of all sampling modules in temporal convolution networks is modified from ref-fcscinetzhang2022less, namely interval sampling and continuous sampling in LightTS zhang2022less, and interactive sampling in ref-fcscinetliu2022scinet.