Comparison of the speed and accuracy of algorithms of optimization of fiber-optic communication lines
Chebykin I.P. 1,2, Starykh D.D. 1, Naniy O.E. 1,3, Andreev G.A.1,2, Treschikov V.N. 1
18 Company, Moscow, Russia
2Moscow Institute of Physics and Technology (National Research University), Dolgoprudny, Moscow Region, Russia
3Moscow State University, Moscow, Russia
Email: chebykin@t8.ru
The study addressed the problem of improvement of the quality of the transmitted optical signal which is measured by the magnitude of the excess of the signal-to-noise ratio over its critical value. It boils down to finding the extremum of a function of many variables unique to each fiber-optic communication line. Various optimization algorithms are compared for finding the most suitable one for this family of functions. The applicability of algorithms based on Bayesian optimization to this problem was demonstrated. The time spent on finding a solution using the proposed algorithms is less than when using deterministic global optimization algorithms, and the deviation from the absolute optimum remains at an acceptable level. Keywords: fiber optic line, signal/noise, optimization algorithm, Bayesian algorithm.
- V.A. Konyshev, A.V. Leonov, O.E. Nanii, A.G. Novikov, V.N. Treshikov, R.R. Ubaydullaev. Quantum Elec., 46 (12), 1121 (2016). DOI: 10.1070/QEL16219
- P. Poggiolini, G. Bosco, A. Carena, V. Curri, Y. Jiang, F. Forghieri. J. Lightwave Technol., 32 (4), 694 (2014). DOI: 10.1109/JLT.2013.2295208
- A. Bononi, R. Dar, M. Secondini, P. Serena, P. Poggiolini. Fiber Nonlinearity and Optical System Performance. In B. Mukherjee, I. Tomkos, M. Tornatore, P. Winzer, Y. Zhao (eds). Springer Handbook of Optical Networks. Springer Handbooks (Springer, Cham., 2020), p. 287-351. DOI: 10.1007/978-3-030-16250-4_9
- A. Carena, G. Bosco, V. Curri, P. Poggiolini, M.T. Taiba, F. Forghieri. Statistical characterization of PM-QPSK signals after propagation in uncompensated fiber links. In 36th European Conference and Exhibition on Optical Communication, IEEE, Sep. 2010, pp. 1-3. DOI: 10.1109/ECOC.2010.5621509
- J.-L. Auge, V. Curri, E. Le Rouzic. Open Design for Multi-Vendor Optical Networks, 2019. [Online]. Available: https://telecominfraproject.com/
- R. Hashemi, H. Beyranvand, H. Rabbani. Opt. Commun., 475, 126212 (2020). DOI: 10.1016/j.optcom.2020.126212
- G. Borraccini, S. Straullu, A. D'Amico, F. Aquilino, S. Piciaccia, A. Tanzi, G. Galimberti, V. Curri. J. Opt. Commun. Networking, 16 (5), B60 (2024). DOI: 10.1364/JOCN.512049
- P. Poggiolini, M.R. Zefreh, G. Bosco, F. Forghieri, S. Piciaccia. Accurate Non-Linearity Fully-Closed-Form Formula based on the GN/EGN Model and Large-Data-Set Fitting. In Optical Fiber Communication Conference (OFC) 2019, Washington, D.C.: OSA, 2019, p. M1I.4. DOI: 10.1364/OFC.2019.M1I.4
- D.D. Starykh, L.A. Samodelkin, O.E. Naniy, V.N. Treschikov. Tech. Phys., 68 (11), 1519 (2023). DOI: 10.61011/TP.2023.11.57504.123-23
- J. Mockus. J. Global Optimization, 4 (4), 347 (1994). DOI: 10.1007/BF01099263
- J. Bergstra, R. Bardenet, Y. Bengio, B. Kegl. Algorithms for Hyper-Parameter Optimization. In J. Shawe-Taylor, R. Zemel, P. Bartlett, F. Pereira, K.Q. Weinberger (eds.). Advances in Neural Information Processing Systems 24 (NIPS 2011)
- D.R. Jones, C.D. Perttunen, B.E. Stuckman. J. Optim. Theory Appl., 79 (1), 157 (1993) [Online]. Available: https://link.springer.com/article/10.1007/BF00941892
Подсчитывается количество просмотров абстрактов ("html" на диаграммах) и полных версий статей ("pdf"). Просмотры с одинаковых IP-адресов засчитываются, если происходят с интервалом не менее 2-х часов.
Дата начала обработки статистических данных - 27 января 2016 г.