An optimized LSTM-based equalizer for 100 Gigabit/s-class short-range fiber-optic communications

Vuong Quang Phuoc; Nguyen Van Dien; Ho Duc Tam Linh; Nguyen Van Tuan; Nguyen Van Hieu; Le Thai Son; Nguyen Tan Hung; Vuong Quang Phuoc; Nguyen Van Dien; Ho Duc Tam Linh; Nguyen Van Tuan; Nguyen Van Hieu; Le Thai Son; Nguyen Tan Hung

doi:10.3934/electreng.2024019

AIMS Electronics and Electrical Engineering

2024, Volume 8, Issue 4: 394-409. doi: 10.3934/electreng.2024019

Previous Article Next Article

Research article Topical Sections

An optimized LSTM-based equalizer for 100 Gigabit/s-class short-range fiber-optic communications

1.
The University of Danang - Advanced Institute of Science and Technology, Danang 50000, Vietnam
2.
The University of Danang - University of Science and Technology, Danang 50000, Vietnam
3.
University of Sciences, Hue University, Hue 49000, Vietnam
4.
FPT University, Danang 50000, Vietnam
5.
Nubis Communications, New Providence NJ 07974, US

Academic Editor: Harry Leib

Received: 28 June 2024 Revised: 20 August 2024 Accepted: 02 September 2024 Published: 09 September 2024

Intensity modulation/direct detection (IM/DD) remains to be the preferred optical transmission scheme for short-range applications for its simplicity of application, inexpensiveness, and small footprint. However, the impairments of low-cost device and fiber chromatic dispersion lead to the limitation of system performance when the data rate rises to 100 Gbps or higher. In this paper, we demonstrated that an equalizer using neural networks can effectively improve the transmission performance of high-speed IM/DD systems. An optimization of a long short-term memory (LSTM) structure in terms of network depth and distribution of neurons in hidden layers leads to an enhancement of the overall performance of the 50 Gbaud PAM4 communications. Furthermore, the results for a system using a LSTM-based equalizer give the better outcome than the traditional feed-forward equalizer (FFE) or artificial neural network (ANN)-based equalizer.
- fiber-optic communications,
- intensity modulation,
- direct detection,
- signal equalization,
- long short-term memory,
- artificial neural network
Citation: Vuong Quang Phuoc, Nguyen Van Dien, Ho Duc Tam Linh, Nguyen Van Tuan, Nguyen Van Hieu, Le Thai Son, Nguyen Tan Hung. An optimized LSTM-based equalizer for 100 Gigabit/s-class short-range fiber-optic communications[J]. AIMS Electronics and Electrical Engineering, 2024, 8(4): 394-409. doi: 10.3934/electreng.2024019

Related Papers:

Abstract

Intensity modulation/direct detection (IM/DD) remains to be the preferred optical transmission scheme for short-range applications for its simplicity of application, inexpensiveness, and small footprint. However, the impairments of low-cost device and fiber chromatic dispersion lead to the limitation of system performance when the data rate rises to 100 Gbps or higher. In this paper, we demonstrated that an equalizer using neural networks can effectively improve the transmission performance of high-speed IM/DD systems. An optimization of a long short-term memory (LSTM) structure in terms of network depth and distribution of neurons in hidden layers leads to an enhancement of the overall performance of the 50 Gbaud PAM4 communications. Furthermore, the results for a system using a LSTM-based equalizer give the better outcome than the traditional feed-forward equalizer (FFE) or artificial neural network (ANN)-based equalizer.

References

[1]	Cisco (2020) Cisco Annual Internet Report (2018–2023) White Paper. Available from: https://www.cisco.com/c/en/us/solutions/collateral/executive-perspectives/annual-internet-report/white-paper-c11-741490.html
[2]	Zhong K, Zhou X, Wang Y, Gui T, Yang Y, Yuan J, et al. (2017) Recent advances in short reach systems. Optical Fiber Communication Conference, Tu2D.7. https://doi.org/10.1364/OFC.2017.Tu2d.7 doi: 10.1364/OFC.2017.Tu2d.7
[3]	Kachris C, Kanonakis K, Tomkos I (2013) Optical interconnection networks in data centers: recent trends and future challenges. IEEE Commun Mag 51: 39‒45. https://doi:10.1109/mcom.2013.6588648 doi: 10.1109/mcom.2013.6588648
[4]	Telecommunication Standardization Sector of ITU, G.989.2: 40-Gigabit-capable passive optical networks 2 (NG-PON2): Physical media dependent (PMD) layer specification, Telecommunication Standardization Sector of ITU 2019. Available from: https://www.itu.int/rec/T-REC-G.989.2
[5]	Telecommunication Standardization Sector of ITU, G.9804.3: 50-Gigabit-capable passive optical networks (50G-PON): Physical media dependent (PMD) layer specification, Telecommunication Standardization Sector of ITU 2021. Available from: https://www.itu.int/rec/T-REC-G.9804.3-202109-I/en
[6]	Wei J, Cheng Q, Penty RV, White IH, Cunningham DG (2015) 400 Gigabit Ethernet using advanced modulation formats: Performance, complexity, and power dissipation. IEEE Commun Mag 53: 182–189. https://doi.org/10.1109/MCOM.2015.7045407 doi: 10.1109/MCOM.2015.7045407
[7]	Zhong K, Zhou X, Gui T, Tao L, Gao Y, Chen W, et al. (2015) Experimental study of PAM-4, CAP-16, and DMT for 100 Gb/s Short Reach Optical Transmission Systems. Opt Express 23: 1176‒1189. https://doi.org/10.1364/OE.23.001176 doi: 10.1364/OE.23.001176
[8]	Zhou H, Li Y, Liu Y, Yue L, Gao C, Li W, et al. (2019) Recent Advances in equalization Technologies for ShortReach Optical Links based on PAM4 modulation: A review. Applied Sciences 9: 2342. https://doi.org/10.3390/app9112342 doi: 10.3390/app9112342
[9]	Stojanovic N, Karinou F, Zhang Q, Prodaniuc C (2017) Volterra and Wiener Equalizers for Short-Reach 100G PAM-4 applications. J Lightwave Technol 35: 4583‒4594. https://doi.org/10.1109/JLT.2017.2752363 doi: 10.1109/JLT.2017.2752363
[10]	Yi L, Tao L, Huang L, Xue L, Li P, Hu W (2019) Machine Learning for 100 Gb/s/λ Passive Optical Network. J Lightwave Technol 37: 1621‒1630. https://doi.org/10.1109/JLT.2018.2888547 doi: 10.1109/JLT.2018.2888547
[11]	Estaran J, Rios-Müller R, Mestre MA, Jorge F, Mardoyan H, Konczykowska A, et al. (2016) Artificial Neural Networks for Linear and Non-Linear Impairment Mitigation in High-Baudrate IM/DD Systems. 42nd European Conference on Optical Communication, 1‒3. VDE.
[12]	Giacoumidis E, Le ST, Aldaya I, Wei JL, McCarthy M, Doran NJ, et al. (2016) Experimental Comparison of Artificial Neural Network and Volterra based Nonlinear Equalization for CO-OFDM. Optical Fiber Communication Conference, W3A-4. https://doi.org/10.1364/OFC.2016.W3A.4 doi: 10.1364/OFC.2016.W3A.4
[13]	Kyono T, Otsuka Y, Fukumoto Y, Owaki S, Nakamura M (2018) Computational Complexity Comparison of Artificial Neural Network and Volterra Series Transfer Function for Optical Nonlinearity Compensation with Time- and Frequency-Domain Dispersion Equalization. European Conference on Optical Communication (ECOC), 1‒3. IEEE. https://doi.org/10.1109/ECOC.2018.8535153
[14]	Hung NT, Stainton S, Le ST, Haigh PA, Tien HP, Vien ND, et al. (2023) High-speed PAM4 transmission using directly modulated laser and artificial neural network nonlinear equaliser. Opt Laser Technol 157: 108642. https://doi.org/10.1016/j.optlastec.2022.108642 doi: 10.1016/j.optlastec.2022.108642
[15]	Schädler M, Böcherer G, Pachnicke S (2021) Soft-Demapping for Short Reach Optical Communication: A Comparison of Deep Neural Networks and Volterra Series. J Lightwave Technol 39: 3095‒3105. https://doi.org/10.1109/JLT.2021.3056869 doi: 10.1109/JLT.2021.3056869
[16]	Nielsen MA (2019) Neural networks and deep learning. Determination Press.
[17]	Goodfellow I, Bengio Y, Courville A (2016) Deep Learning. MIT Press, 800 pp.
[18]	Dai X, Li X, Luo M, You Q, Yu S (2019) LSTM networks enabled nonlinear equalization in 50-Gb/s PAM-4 transmission links. Appl Optics 58: 6079‒6084. https://doi.org/10.1364/AO.58.006079 doi: 10.1364/AO.58.006079
[19]	Peng CW, Chan DW, Chow CW, Hung TY, Jian YH, Tong Y, et al. (2023) Long Short Term Memory Neural Network (LSTMNN) and inter-symbol feature extraction for 160 Gbit/s PAM4 from silicon micro-ring transmitter. Opt Commun 529: 129067. https://doi.org/10.1016/j.optcom.2022.129067 doi: 10.1016/j.optcom.2022.129067
[20]	Wu Q, Xu Z, Zhu Y, Zhang Y, Ji H, Yang Y, et al. (2023) Machine learning for Self-Coherent detection Short-Reach optical communications. Photonics 10: 1001. https://doi.org/10.3390/photonics10091001 doi: 10.3390/photonics10091001
[21]	Hochreiter S, Schmidhuber J (1997) Long Short-Term memory. Neural Comput 9: 1735‒1780. https://doi.org/10.1162/neco.1997.9.8.1735 doi: 10.1162/neco.1997.9.8.1735
[22]	Jozefowicz R, Zaremba W, Sutskever I (2015) An empirical exploration of recurrent network architectures. Proceedings of the 32nd International Conference on Machine Learning, 2342‒2350. PMLR.
[23]	Telecommunication Standardization Sector of ITU. G.694.2 : Spectral Grids for WDM Applications: CWDM Wavelength Grid. Telecommunication Standardization Sector of ITU 2003. Available from: https://www.itu.int/rec/T-REC-G.694.2-200312-I/en

Reader Comments

Your name:*

Email:*
© 2024 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)