[MB1] Nonlinear Transmission

[MB1-2] Dynamic complex deep Neural Network Nonlinear Equalizer for 64 QAM Long-haul Transmission Systems

〇Govind Sharan Yadav¹, Takehiro Tsuritani², Shohei Beppu², Hidenori Takahashi², Itsuro Morita², Kai-Ming Feng¹, Jhih-heng Yan¹ (¹National Tsing Hua Univ., Taiwan, ²KDDI Research Inc., Japan)

キーワード：Digital signal processing techniques for optical communications , Complex Deep Artificial Neural Network, Deep Learning, Adaptive machine learning method

We implemented a two-hidden-layer dynamic complex deep neural network nonlinear equalizer which outperforms the linear equalizer, static and dynamic single hidden layer CDNN-NLE by 1.38-dB, 1.01-dB and 0.62-dB for a 34-GBaud/s, 64-QAM signal over 1200-km.