Demodulation of optical eigenvalue modulated signal using neural network

Shohei Yamamoto; Ken Mishina; Akihiro Maruta

doi:10.1587/comex.2019GCL0026

This article has now been updated. Please use the final version.

Demodulation of optical eigenvalue modulated signal using neural network

Shohei Yamamoto, Ken Mishina, Akihiro Maruta

Author information

Keywords: Optical fiber communication, machine learning, multi-soliton transmission, optical eigenvalue modulation, neural network

JOURNAL FREE ACCESS Advance online publication

Article ID: 2019GCL0026

DOI https://doi.org/10.1587/comex.2019GCL0026

The final version of this article is now available: Vol. 8 (2019), No. 12 pp. 507-512

Details

Abstract

Eigenvalue communication is a promising technology for overcoming capacity limitation due to Kerr nonlinearity in optical fiber. In this letter, we propose an artificial neural network based demodulation method for optical eigenvalue modulated signal using on-off encoding. The proposed method has 9.2 dB power margin at BER = 3.8×10³ compared with the conventional inverse scattering transform based demodulation method.

Corresponding author

Register with J-STAGE for free!