Photonic-assisted Modulation Format Identification Using Convolutional Neural Networks

A photonic-assisted architecture is used to implement self-interference on radio-frequency (RF) signals to create modulation-biased amplitude patterns. The interfered signals are identified from their modulation formats using a convolutional neural network (CNN). In the proof-of-concept experiments, the average accuracy reaches 97.06% for signals with 10 modulation formats at 1 G Baud.