Modern statistical learning technologies such as deep learning have a great potential to deal with linear/nonlinear fiber impairments for future coherent optical communications. We introduce various learning techniques suited for nonlinear equalizations.
© 2014 OSA
T. Koike-Akino, "Perspective of Statistical Learning for Nonlinear Equalization in Coherent Optical Communications," in Advanced Photonics for Communications, OSA Technical Digest (online) (Optical Society of America, 2014), paper ST2D.2.
References are not available for this paper.