Related Experiment Video
Updated: May 21, 2026
09:43
Transmission of Multiple Signals through an Optical Fiber Using Wavefront Shaping
Published on: March 20, 2017
Modulation format identification in heterogeneous fiber-optic networks using artificial neural networks.
Faisal Nadeem Khan1, Yudi Zhou, Alan Pak Tao Lau
1Photonics Research Centre, The Hong Kong Polytechnic University, Hung Hom, Kowloon, Hong Kong China. eefaisal.nadeem@eng.usm.my
Optics Express
|June 21, 2012
Summary
This study introduces a simple, cost-effective method for modulation format identification in fiber-optic networks using artificial neural networks and asynchronous amplitude histograms. The technique achieves 99.6% accuracy, even with network impairments.
Area of Science:
- Optical Communications
- Artificial Intelligence
- Signal Processing
Background:
- Next-generation fiber-optic networks require efficient methods for managing diverse modulation formats.
- Accurate modulation format identification (MFI) is crucial for network performance and adaptability.
Purpose of the Study:
- To develop a simple, cost-effective MFI technique for heterogeneous fiber-optic networks.
- To utilize artificial neural networks (ANNs) trained on asynchronous amplitude histogram (AAH) features for MFI.
Main Methods:
- Feature extraction from asynchronous amplitude histograms (AAHs).
- Training an artificial neural network (ANN) using the extracted AAH features.
- Numerical simulations for MFI across six modulation formats and varying data rates.
Main Results:
- The proposed ANN-based technique achieved an overall estimation accuracy of 99.6% for MFI.
- Effective classification was demonstrated across six common modulation formats.
- The technique proved robust in the presence of various link impairments.
Conclusions:
- The proposed simple hardware and digital signal processing (DSP) technique enables effective MFI.
- This adaptable method can identify various modulation formats at different data rates without hardware modifications.
