We demonstrate a technique of using artificial neural networks for optical performance monitoring of PSK data signals. Parameters for training are derived from delay-tap asynchronous diagrams using balanced detection. We also compare the results with the case of using direct detection.
Proceedings Title: 35th European Conference on Optical Communications
Conference Dates: September 20-24, 2009
Conference Location: Vienna, -1
Pub Type: Conferences
artificial neural network, asynchronous sampling, balanced detection, optical performance monitoring