Reinforcement Learning for Generalized Parameter Optimization in Elastic Optical Networks

被引：11

作者：

Koch, Rebekka ^{[1
]}

Kuehl, Sebastian ^{[1
]}

Morais, Rui Manuel ^{[2
,3
]}

Spinnler, Bernhard ^{[4
]}

Schairer, Wolfgang ^{[4
]}

Sommernkorn-Krombholz, Bernd ^{[5
,6
]}

Pachnicke, Stephan ^{[1
]}

机构：

[1] Univ Kiel, Chair Commun, D-24143 Kiel, Germany

[2] Infinera Portugal, P-2790078 Carnaxide, Portugal

[3] BNP Paribas CIB, Lisbon, Portugal

[4] Infinera Germany, D-81541 Munich, Germany

[5] Infinera, D-81541 Munich, Germany

[6] ADVA Opt Networking SE, D-82152 Munich, Germany

来源：

JOURNAL OF LIGHTWAVE TECHNOLOGY | 2022年 / 40卷 / 03期

关键词：

Optical fiber networks; Genetic algorithms; Optimization; Optical fiber amplifiers; Statistics; Sociology; Optical fiber dispersion; Advantage actor-critic algorithm; deep reinforcement learning; elastic optical networks (EONs); modulation and spectrum assignment (RMSA); routing; GN MODEL; ALLOCATION;

D O I：

10.1109/JLT.2021.3123271

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

0808 ; 0809 ;

摘要：

Elastic Optical Networking enables efficient use of spectral resources at the cost of a large parameter space which needs to be optimized to maximize transmission bandwidth. By formulating this optimization problem as a Markov Decision Process, we show that by using a state of the art Reinforcement Learning algorithm an agent can be trained, which is able to select near optimal parameters for different link conditions within seconds. Furthermore, the trained agent is able to generalize to unseen conditions, removing the need to optimize and train for every possible link scenario.

引用

页码：567 / 574

页数：8

共 19 条

[1] Ali H, 2019, ARXIV190801275
[2] EGN model of non-linear fiber propagation
Carena, Andrea
Bosco, Gabriella
Curri, Vittorio
Jiang, Yanchao
Poggiolini, Pierluigi
Forghieri, Fabrizio
[J]. OPTICS EXPRESS, 2014, 22 (13): : 16335 - 16362
[3] Chen XF, 2018, IEEE VTS VEH TECHNOL
[4] Degris T, 2012, P AMER CONTR CONF, P2177
[5] A Survey of Actor-Critic Reinforcement Learning: Standard and Natural Policy Gradients
Grondman, Ivo
Busoniu, Lucian
Lopes, Gabriel A. D.
Babuska, Robert
[J]. IEEE TRANSACTIONS ON SYSTEMS MAN AND CYBERNETICS PART C-APPLICATIONS AND REVIEWS, 2012, 42 (06): : 1291 - 1307
[6] Hill A., 2018, Stable baselines
[7] Ives DJ, 2013, 2013 OPTICAL FIBER COMMUNICATION CONFERENCE AND EXPOSITION AND THE NATIONAL FIBER OPTIC ENGINEERS CONFERENCE (OFC/NFOEC)
[8] Mitchell M., 1998, An Introduction to Genetic Algorithms. Complex Adaptive Systems
[9] Mnih V, 2016, PR MACH LEARN RES, V48
[10] Poggiolini P., 2015, ARXIV150304132

← 1 2 →