Study of Training Patterns for Employing Deep Neural Networks in Optical Communication Systems
Our investigation demonstrates that training patterns generated by Mersenne Twister pseudo-random number generator have potential to overcome the pattern effect. The resulting trained networks are capable of compensating the actual optical communication system response rather than merely predicting training signals.
Paper
Full text
Study of Training Patterns for Employing Deep Neural Networks in Optical Communication Systems
Semantic Scholar · Engineering · 2018
Abstract
Our investigation demonstrates that training patterns generated by Mersenne Twister pseudo-random number generator have potential to overcome the pattern effect. The resulting trained networks are capable of compensating the actual optical communication system response rather than merely predicting training signals.