[1] [1] CHENG W, SONG H, LI D, et al. A robust sparse RLS-volterra nonlinear equalizer using l0-regularization for 4×150 Gbit/s IMDD-based optical interconnect[J]. IEEE Access, 2021, 9: 30881-30892.

[3] [3] WU Q, ZHU Y, HU W. Piecewise feedforward neural network based nonlinear equalizer for short-reach DML-DD system[C]// 2021 19th Inter-national Conference on Optical Communications and Networks, August 23-27, 2021, Qufu, China. New York: IEEE, 2021: 1-3.

[4] [4] HOSSAIN S B, RAHMAN T, STOJANOVIC N, et al. Transmission beyond 200 Gbit/s with IM/DD system for campus and intra-datacenternet-work applications[J]. IEEE Photonics Technology Letters, 2021, 33(5): 263-266.

[6] [6] ZHANG K, ZHUGE Q, XIN H, et al. Intensity directed equalizer for the mitigation of DML chirp induced distortion in dispersion-unmanaged C-band PAM transmission[J]. Optics Express, 2017, 23: 28123-28135.

[7] [7] PANG X, UDALCOVS A, SCHATZ R, et al. High-speed short reach optical communications: technological options and challenges[C]//2020 Asia Communications and Photonics Conference and International Confe-rence on Information Photonics and Optical Communications, October 24-27, 2020, Beijing, China. New York: IEEE, 2020: 1-3.

[8] [8] GAO F, ZHOU S, LI X, et al. 2×64 Gb/s PAM-4 transmission over 70 km SSMF using O-band 18G-class directly modulated lasers (DMLs)[J]. Opti-cs Express, 2017, 7: 7230-7237.

[9] [9] ZHAI Z, JIANG H, FU M, et al. An interpretable mapping from a com-munication system to a neural network for optimal transceiver-joint equa-lization[J]. Journal of Lightwave Technology, 2021, 39(17): 5449-5458.

[10] [10] SANG B, ZHANG J, WANG C, et al. Multi-symbol output long short-term memory neural network equalizer for 200+Gbps IM/DD system[C]//2021 European Conference on Optical Communication, September 13-16, 2021, Bordeaux, France. New York: IEEE, 2021: 1-4.

[11] [11] ZHANG J, YAN L, JIANG L, et al. Convolutional neural network equalizer for Short-reach optical communication systems[C]//Asia Com-munications and Photonics Conference, October 24-27, 2020, Beijing, China. New York: IEEE, 2020: 1-3.

[12] [12] XU Z, SUN C, JI T, et al. Investigation on the impact of additional con-nections to feedforward neural networks for equalization in PAM4 short-reach direct detection links[C]//Asia Communications and Photonics Con-ference. October 24-27, 2020, Beijing, China. New York: IEEE, 2020: 1-3.

[13] [13] LUO M, GAO F, LI X, et al. Transmission of 4×50Gb/s PAM-4 signal over 80-km single mode fiber using neural network[C]//Optical Fiber Com-munication Conference. March 11-15, 2018, San Diego, California United States. New York: IEEE, 2018: 1-3.

[14] [14] LI P, YI L, XUE L, et al. 56 Gbps IM/DD PON based on 10G-class optical devices with 29 dB loss budget enabled by machine learning[C]//2018 Optical Fiber Communications Conference and Exposition. March 11-15, 2018, San Diego, California United States. New York: IEEE, 2018: 1-3.

[15] [15] LIN X, LUO S, SOAMN S K O, et al. Perturbation theory-aided learned digital back-propagation scheme for optical fiber nonlinearity com-pensation[J]. Journal of Lightwave Technology, 2021, 40(7): 1981-1988.

[16] [16] SZE V, CHEN Y, YANG T, et al. Efficient processing of deep neural networks: a tutorial and survey[J]. Proceedings of the IEEE, 2017, 105(12): 2295-2329.

[17] [17] XI Y, BI M, XIN M, et al. Simplified SVM equalization algorithm based on single hyperplane training enabled 50Gb/s PAM-4/8 with 10-G optics in NG-PON system[J]. IEEE Photonics Journal, 2021, 13(4): 1-7.