论文信息 - 800-Gbit/s/carrier TPS-64QAM WDM Coherent Transmission over 2,400 km Utilizing Low-complexity Separated Pruning DNN-based Nonlinear Equalization

800-Gbit/s/carrier TPS-64QAM WDM Coherent Transmission over 2,400 km Utilizing Low-complexity Separated Pruning DNN-based Nonlinear Equalization

We experimentally demonstrated 800-Gbit/s/carrier WDM coherent transmission over 2,400-km based on 100-GBd truncated PS-64QAM utilizing 75%-sparsity pruning DNN-based nonlinear equalization. Results show that our pruning DNN-NLE with 24%-lower complexity outperforms Volterra NLE by 20% reach improvement.

[1] Wen Zhou,et al. 800-Gb/s/carrier WDM Coherent Transmission Over 2000 km Based on Truncated PS-64QAM Utilizing MIMO Volterra Equalizer , 2022, Journal of Lightwave Technology.

[2] 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.

[3] Kaihui Wang,et al. Mitigation of SOA-Induced Nonlinearity With the Aid of Deep Learning Neural Networks , 2022, Journal of Lightwave Technology.

[4] M. Chagnon,et al. Deep Neural Network-Based Digital Pre-Distortion for High Baudrate Optical Coherent Transmission , 2022, Journal of Lightwave Technology.

[5] Wen Zhou,et al. 81-GHz W-band 60-Gbps 64-QAM wireless transmission based on a dual-GRU equalizer. , 2022, Optics Express.

[6] Toshiaki Koike-Akino,et al. Zero-Multiplier Sparse DNN Equalization for Fiber-Optic QAM Systems with Probabilistic Amplitude Shaping , 2021, 2021 European Conference on Optical Communication (ECOC).

[7] 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.

[8] Weisheng Hu,et al. Machine Learning for 100 Gb/s/λ Passive Optical Network , 2019, Journal of Lightwave Technology.

[9] Song Han,et al. Deep Compression: Compressing Deep Neural Network with Pruning, Trained Quantization and Huffman Coding , 2015, ICLR.