Machine Learning based Equalization

Th.A.3.1

Invited Speaker  - Low-Complexity Efficient Neural Network Optical Channel Equalizers: Training, Inference, and Hardware Synthesis

Pedro J. Freire1, Sasipim Srivallapanondh1, Bernhard Spinnler2, Antonio Napoli2, Nelson Costa2, Jaroslaw E. Prilepsky1, Sergei K. Turitsyn1 1 Aston University, Birmingham, United Kingdom. 2 Infinera, Munich, Germany

Th.A.3.2

Area-Efficient Hardware Parallelization of Neural Network CD Equalizers for 4×200 Gb/s PAM4 CWDM4 Systems

Bo Liu1, Christian Bluemm2, Stefano Calabrò2, Bing Li1, Ulf Schlichtmann1 1 Technical University of Munich, Chair for Electronic Design Automation, Munich, Germany. 2 Huawei Technologies Duesseldorf GmbH, Munich, Germany

Th.A.3.3

Blind Frequency-Domain Equalization Using Vector-Quantized Variational Autoencoders

Jinxiang Song1, Vincent Lauinger2, Christian H\"{a}ger1, Jochen Schr\"{o}der1, Alexandre Graell i Amat1, Laurent Schmalen2, Henk Wymeersch1

1 Chalmers University of Technology, Gothenburg, Sweden. 2 Karlsruhe Institute of Technology, Karlsruhe, Germany

Th.A.3.4

Mixed-Precision Integer-Arithmetic-Only Neural Network-Based Equalizers for DML-Based Short-Reach IM/DD Systems

Zhaopeng Xu1, Honglin Ji1, Yu Yang1, Gang Qiao1, Qi Wu1, Jia Li1, Weiqi Lu2, Lulu Liu1, Shangcheng Wang1, Jinlong Wei1, Zhixue He1, Weisheng Hu1, William Shieh2 1 Peng Cheng Laboratory, Shenzhen, China. 2 Westlake University, Hangzhou, China

Th.A.3.5

140-Gbaud PAM-8 IM/DD Transmission and FTN Signal Processing based on Low-Complexity Nonlinear M-BCJR Equalization with Deep Neural-Network Channel Model

An Yan1, Sizhe Xing1, Guoqiang Li1, Zhongya Li1, Penghao Luo1, Aolong Sun1, Jianyang Shi1, Hongguang Zhang2, Xi Xiao2, Zhixue He3, Nan Chi1, Junwen Zhang1 1 Fudan University, Shanghai, China. 2 National Information Optoelectronics Innovation Center, Wuhan, China. 3 Peng Cheng Lab, Shenzhen, China