Chinese Optics Letters, Volume. 21, Issue 3, 031901(2023)
Characteristic extraction of soliton dynamics based on convolutional autoencoder neural network
Figures & Tables(7)
Fig. 2. Double solitons collision. (a) Time-domain evolution; (b) spectrum.
Fig. 3. Double solitons. (a) Reconstructed spectra; (b) PCCs; (c) reconstructed field autocorrelation trajectory; (d) soliton separation and relative phase of the reconstructed 6000th round trip; (e) soliton separation and relative phase of the original 6000th round trip.
Fig. 4. Double solitons. (a) Relationship between the loss of double soliton collision and number of dense n; (b) latent parameters.
Fig. 5. Triple solitons collision. (a) Time-domain evolution; (b) spectrum.
Fig. 6. Triple solitons. (a) Reconstructed spectra; (b) PCCs; (c) reconstructed field autocorrelation trajectory; (d) separation and relative phase of the reconstructed 1680th round trip; (e) separation and relative phase of the original 1680th round trip.
Fig. 7. Triple solitons. (a) Relationship between the loss of double soliton collision and number of dense n; (b) latent parameters.
Tools
Get Citation
Copy Citation Text
Congcong Liu, Jiangyong He, Pan Wang, Dengke Xing, Jin Li, Yange Liu, Zhi Wang, "Characteristic extraction of soliton dynamics based on convolutional autoencoder neural network," Chin. Opt. Lett. 21, 031901 (2023)
Paper Information
Category: Nonlinear Optics
Received: Jul. 8, 2022
Accepted: Sep. 20, 2022
Posted: Sep. 21, 2022
Published Online: Nov. 7, 2022
The Author Email: Jiangyong He (jiangyonghe@nankai.edu.cn), Zhi Wang (zhiwang@nankai.edu.cn)
© Copyright 2018-2021 | Chinese Laser Press.
All Rights Reserved 沪ICP备15018463号-20