We demonstrate a technique of using artificial neural networks trained with parameters derived from delay-tap asynchronous diagrams for optical performance monitoring of phase shift keying (PSK) data signals. We show that asynchronous diagrams from balanced detection give superior results compared to direct detection in a 40 Gb/s binary PSK system. Experimental demonstration in a 100 Gb/s quadrature PSK system verifies the effectiveness of the proposed technique.
Citation: IEEE Photonics Technology Letters
Pub Type: Journals
neural networks, optical fiber communication, optical performance monitoring, phase modulation