On the Computational Complexity of Artificial Neural Networks for Short-Reach Optical Communication

Artificial neural networks (ANNs) have been widely used for nonlinear equalization in short-reach optical communications due to the superior performance compared with traditional approaches. However, the computational complexity (CC) is a major challenge that hinder their applications. In this paper, we compare various types of ANNs for short-reach optical communication with CC constraint, as well as provide several means of CC reduction by employing transfer learning, pruning and multi-task networks.

[1]  Govind Sharan Yadav,et al.  Sparsity Learning Deep Neural Network Nonlinear Equalization Method for 112Gbps PAM4 Short-Reach Transmission Links , 2023, Journal of Lightwave Technology.

[2]  M. Kuschnerov,et al.  Spiking Neural Network Linear Equalization: Experimental Demonstration of 2km 100Gb/s IM/DD PAM4 Optical Transmission , 2023, 2023 Optical Fiber Communications Conference and Exhibition (OFC).

[3]  M. Kuschnerov,et al.  800Gb/s PAM4 Transmission over 10km SSMF Enabled by Low-complex Duobinary Neural Network Equalization , 2022, 2022 European Conference on Optical Communication (ECOC).

[4]  Lin Sun,et al.  Optimizations and investigations for transfer learning of iteratively pruned neural network equalizers for data center networking. , 2022, Optics Express.

[5]  Gordon Ning Liu,et al.  Fastly converged transfer learning using neuron-pruning nonlinear equalizer for intra data center networking , 2022, 2022 27th OptoElectronics and Communications Conference (OECC) and 2022 International Conference on Photonics in Switching and Computing (PSC).

[6]  M. Kuschnerov,et al.  Spiking Neural Network Equalization on Neuromorphic Hardware for IM/DD Optical Communication , 2022, 2022 European Conference on Optical Communication (ECOC).

[7]  Wen Zhou,et al.  Low Complexity Neural Network Equalization Based on Multi-Symbol Output Technique for 200+ Gbps IM/DD Short Reach Optical System , 2022, Journal of Lightwave Technology.

[8]  J. Manton,et al.  Computationally-Efficient Sparsely-Connected Multi-Output Neural Networks for IM/DD System Equalization , 2022, 2022 Optical Fiber Communications Conference and Exhibition (OFC).

[9]  Xiaofeng Hu,et al.  Low-Complexity Recurrent Neural Network Based Equalizer With Embedded Parallelization for 100-Gbit/s/λ PON , 2022, Journal of Lightwave Technology.

[10]  Pedro J. Freire,et al.  Domain Adaptation: the Key Enabler of Neural Network Equalizers in Coherent Optical Systems , 2022, 2022 Optical Fiber Communications Conference and Exhibition (OFC).

[11]  John C Manton,et al.  Low-Complexity Multi-Task Learning Aided Neural Networks for Equalization in Short-Reach Optical Interconnects , 2022, Journal of Lightwave Technology.

[12]  W. Shieh,et al.  Carrier Assisted Differential Detection With Generalized and Simplified Receiver Structure , 2021, Journal of Lightwave Technology.

[13]  Zuyuan He,et al.  High-throughput hardware deployment of pruned neural network based nonlinear equalization for 100-Gbps short-reach optical interconnect. , 2021, Optics Letters.

[14]  J. Manton,et al.  Sparsely-Connected Cascade Recurrent Neural Network-Based Nonlinear Equalizer for a 100-Gb/s PAM4 Optical Interconnect , 2021, 2021 Asia Communications and Photonics Conference (ACP).

[15]  Jian Zhao,et al.  Transfer-Learning Based Convolutional Neural Network for Short-Reach Optical Interconnects , 2021, Asia Communications and Photonics conference and Exhibition.

[16]  Miao Kong,et al.  Multi-Symbol Output Long Short-Term Memory Neural Network Equalizer For 200+ Gbps IM/DD System , 2021, 2021 European Conference on Optical Communication (ECOC).

[17]  W. Pan,et al.  56-Gbit/s PAM-4 Optical Signal Transmission Over 100-km SMF Enabled by TCNN Regression Model , 2021, IEEE Photonics Journal.

[18]  J. Manton,et al.  Towards Low Computational Complexity for Neural Network-Based Equalization in PAM4 Short-Reach Direct Detection Systems by Multi-Symbol Prediction , 2021, Optical Fiber Communications Conference and Exhibition.

[19]  Chi-Wai Chow,et al.  Long Short-Term Memory Neural Network for Mitigating Transmission Impairments of 160 Gbit/s PAM4 Microring Modulation , 2021, 2021 Optical Fiber Communications Conference and Exhibition (OFC).

[20]  Xiaofeng Hu,et al.  Recurrent Neural Network based Equalizer with Embedded Parallelization for 100Gbps/λ PON , 2021, 2021 Optical Fiber Communications Conference and Exhibition (OFC).

[21]  J. Manton,et al.  Joint Equalization of Linear and Nonlinear Impairments for PAM4 Short-Reach Direct Detection Systems , 2021, IEEE Photonics Technology Letters.

[22]  Govind Sharan Yadav,et al.  Computation efficient sparse DNN nonlinear equalization for IM/DD 112  Gbps PAM4 inter-data center optical interconnects. , 2021, Optics letters.

[23]  Pedro J. Freire,et al.  Transfer Learning for Neural Networks-Based Equalizers in Coherent Optical Systems , 2021, Journal of Lightwave Technology.

[24]  J. Manton,et al.  Feedforward and Recurrent Neural Network-Based Transfer Learning for Nonlinear Equalization in Short-Reach Optical Links , 2021, Journal of Lightwave Technology.

[25]  Zhaopeng Xu,et al.  Theoretical and Experimental Investigations of Interleaved Carrier-Assisted Differential Detection , 2021, Journal of Lightwave Technology.

[26]  W. Shieh,et al.  Investigation on the Impact of Additional Connections to Feedforward Neural Networks for Equalization in PAM4 Short-Reach Direct Detection Links , 2020, 2020 Asia Communications and Photonics Conference (ACP) and International Conference on Information Photonics and Optical Communications (IPOC).

[27]  Zhaopeng Xu,et al.  Cascade recurrent neural network-assisted nonlinear equalization for a 100  Gb/s PAM4 short-reach direct detection system. , 2020, Optics letters.

[28]  Zhaopeng Xu,et al.  Experimental Demonstration of Complex-Valued DSB Signal Field Recovery via Direct Detection , 2020, IEEE Photonics Technology Letters.

[29]  Zuyuan He,et al.  Compressed Neural Network Equalization Based on Iterative Pruning Algorithm for 112-Gbps VCSEL-Enabled Optical Interconnects , 2020, Journal of Lightwave Technology.

[30]  Weisheng Hu,et al.  Neural Network-Based Equalization in High-Speed PONs , 2020, 2020 Optical Fiber Communications Conference and Exhibition (OFC).

[31]  W. Shieh,et al.  Cascade Recurrent Neural Network Enabled 100-Gb/s PAM4 Short-Reach Optical Link Based on DML , 2020, 2020 Optical Fiber Communications Conference and Exhibition (OFC).

[32]  Keren Bergman,et al.  FPGA Implementation of Deep Neural Network Based Equalizers for High-Speed PON , 2020, 2020 Optical Fiber Communications Conference and Exhibition (OFC).

[33]  Fan Yang,et al.  160-Gb/s Nyquist PAM-4 Transmission with GeSi-EAM using Artificial Neural Network Based Nonlinear Equalization , 2020, 2020 Optical Fiber Communications Conference and Exhibition (OFC).

[34]  Massimo Tornatore,et al.  Active vs Transfer Learning Approaches for QoT Estimation with Small Training Datasets , 2020, 2020 Optical Fiber Communications Conference and Exhibition (OFC).

[35]  Zhaopeng Xu,et al.  Transfer Learning Aided Neural Networks for Nonlinear Equalization in Short-Reach Direct Detection Systems , 2020, 2020 Optical Fiber Communications Conference and Exhibition (OFC).

[36]  Weisheng Hu,et al.  200 Gbps/Lane IM/DD Technologies for Short Reach Optical Interconnects , 2020, Journal of Lightwave Technology.

[37]  Chuanbowen Sun,et al.  Computational complexity comparison of feedforward/radial basis function/recurrent neural network-based equalizer for a 50-Gb/s PAM4 direct-detection optical link. , 2019, Optics express.

[38]  Min Zhang,et al.  Comprehensive Eye Diagram Analysis: A Transfer Learning Approach , 2019, IEEE Photonics Journal.

[39]  Zhaopeng Xu,et al.  Computational Complexity Analysis of Neural Network-Based Nonlinear Equalization for Short Reach Direct Detection Systems , 2019, 2019 Asia Communications and Photonics Conference (ACP).

[40]  Jing Zhang,et al.  Fast remodeling for nonlinear distortion mitigation based on transfer learning. , 2019, Optics letters.

[41]  Shaohua Yu,et al.  LSTM networks enabled nonlinear equalization in 50-Gb/s PAM-4 transmission links. , 2019, Applied optics.

[42]  Yingxiong Song,et al.  Transfer learning assisted deep neural network for OSNR estimation. , 2019, Optics express.

[43]  Zuyuan He,et al.  Threshold-Based Pruned Retraining Volterra Equalization for 100 Gbps/Lane and 100-m Optical Interconnects Based on VCSEL and MMF , 2019, Journal of Lightwave Technology.

[44]  Zhangyuan Chen,et al.  Low computationally complex recurrent neural network for high speed optical fiber transmission , 2019, Optics Communications.

[45]  Chia-Chien Wei,et al.  Sparse Volterra Nonlinear Equalizer by Employing Pruning Algorithm for High-Speed PAM-4 850-nm VCSEL Optical Interconnect , 2019, 2019 Optical Fiber Communications Conference and Exhibition (OFC).

[46]  Polina Bayvel,et al.  End-to-End Optimized Transmission over Dispersive Intensity-Modulated Channels Using Bidirectional Recurrent Neural Networks , 2019, Optics express.

[47]  Xiaofeng Hu,et al.  Recurrent Neural Network (RNN) Based End-to-End Nonlinear Management for Symmetrical 50Gbps NRZ PON with 29dB+ Loss Budget , 2018, 2018 European Conference on Optical Communication (ECOC).

[48]  Ahmed Galib Reza,et al.  Nonlinear Equalizer Based on Neural Networks for PAM-4 Signal Transmission Using DML , 2018, IEEE Photonics Technology Letters.

[49]  Deming Liu,et al.  Radial basis function neural network enabled C-band 4 × 50  Gb/s PAM-4 transmission over 80  km SSMF. , 2018, Optics letters.

[50]  Polina Bayvel,et al.  End-to-End Deep Learning of Optical Fiber Communications , 2018, Journal of Lightwave Technology.

[51]  Songnian Fu,et al.  Transmission of 4×50-Gb/s PAM-4 Signal over 80-km Single Mode Fiber using Neural Network , 2018, 2018 Optical Fiber Communications Conference and Exposition (OFC).

[52]  Chia-Chien Wei,et al.  Convolutional Neural Network based Nonlinear Classifier for 112-Gbps High Speed Optical Link , 2018, 2018 Optical Fiber Communications Conference and Exposition (OFC).

[53]  Weisheng Hu,et al.  100Gbps IM/DD Transmission over 25km SSMF using 20G-class DML and PIN Enabled by Machine Learning , 2018, 2018 Optical Fiber Communications Conference and Exposition (OFC).

[54]  Chao Lu,et al.  Digital Signal Processing for Short-Reach Optical Communications: A Review of Current Technologies and Future Trends , 2018, Journal of Lightwave Technology.

[55]  Hanlin Feng,et al.  Demonstration of 50Gbps IM/DD PAM4 PON over 10GHz Class Optics Using Neural Network Based Nonlinear Equalization , 2017, 2017 European Conference on Optical Communication (ECOC).

[56]  Liang Zhang,et al.  IM/DD Transmission Techniques for Emerging 5G Fronthaul, DCI, and Metro Applications , 2017, Journal of Lightwave Technology.

[57]  Chao Yang,et al.  IM/DD-Based 112-Gb/s/lambda PAM-4 Transmission Using 18-Gbps DML , 2016, IEEE Photonics Journal.

[58]  Ali Ghiasi,et al.  Large data centers interconnect bottlenecks. , 2015, Optics express.

[59]  Francesco Da Ros,et al.  Reservoir-computing and neural-network-based equalization for short reach communication , 2022, Optica Advanced Photonics Congress 2022.

[60]  William Shieh,et al.  High-capacity direct-detection systems , 2020 .

[61]  Chia-Chien Wei,et al.  Employing Deep Neural Network for High Speed 4-PAM Optical Interconnect , 2017, 2017 European Conference on Optical Communication (ECOC).