Abstract

We demonstrate a machine-learning-assisted modulation format identification scheme using both intensity and differential-phase density information. The identification performance for all formats can be achieved even at OSNR values lower than corresponding theoretical 20% FEC limit.

© 2020 The Author(s)

PDF Article

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