A new deep learning training method for digital back propagation (DBP) is introduced. It is invariant to polarization state rotation and phase noise. Applying the method one gains more than 1 dB over standard DBP.
© 2020 The Author(s)
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
Login to access OSA Member Subscription