Abstract

We have proposed and demonstrated a transfer learning (TL)-assisted deep learning network (DNN) for nonlinear distortion compensation in optical side-band PAM-4 modulation and direct-detection transmission. Since there exists partial correlation of nonlinear distortions, we can transfer the parameters of the trained DNN to the target model to speed up remodeling and reduce complexity. We conduct experiments to demonstrate the effectiveness of the proposed scheme in Nyquist PAM-4 transmissions. The required iterations or train size with TL can be less than half of that with retraining without any performance degradation in single-channel transmission. We also extend the proposed scheme into multi-channel transmissions. Since all channels co-propagate on the same fiber, the correlation of nonlinear distortions enables TL to share the parameters among different channels and realize fast remodeling with a better starting rather than a stochastic beginning. The experimental results show that the iterations of TL are one-fourth that of retraining without performance penalty in five-channel transmission.

© 2019 Optical Society of America

Full Article  |  PDF Article
OSA Recommended Articles
Transfer learning assisted deep neural network for OSNR estimation

Le Xia, Jing Zhang, Shaohua Hu, Mingyue Zhu, Yingxiong Song, and Kun Qiu
Opt. Express 27(14) 19398-19406 (2019)

Gaussian kernel-aided deep neural network equalizer utilized in underwater PAM8 visible light communication system

Nan Chi, Yiheng Zhao, Meng Shi, Peng Zou, and Xingyu Lu
Opt. Express 26(20) 26700-26712 (2018)

Comparison of digital signal-signal beat interference compensation techniques in direct-detection subcarrier modulation systems

Zhe Li, M. Sezer Erkilinc, Lidia Galdino, Kai Shi, Benn C. Thomsen, Polina Bayvel, and Robert I. Killey
Opt. Express 24(25) 29176-29189 (2016)

References

You do not have subscription access to this journal. Citation lists with outbound citation links are available to subscribers only. You may subscribe either as an OSA member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access OSA Member Subscription

Cited By

You do not have subscription access to this journal. Cited by links are available to subscribers only. You may subscribe either as an OSA member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access OSA Member Subscription

Figures (6)

You do not have subscription access to this journal. Figure files are available to subscribers only. You may subscribe either as an OSA member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access OSA Member Subscription

Equations (2)

You do not have subscription access to this journal. Equations are available to subscribers only. You may subscribe either as an OSA member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access OSA Member Subscription

Metrics

You do not have subscription access to this journal. Article level metrics are available to subscribers only. You may subscribe either as an OSA member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access OSA Member Subscription