Dynamic power pre-adjustments with machine learning that mitigate EDFA excursions during defragmentation

We examine EDFA power excursions during three defragmentation methods of flexgrid super-channels. Using a machine learning approach, we demonstrate automatic and dynamic adjustments of pre-EDFA power levels, and show the mitigation of post-EDFA power discrepancy among channels by over 62%.

[1]  S. J. B. Yoo,et al.  Spectrum defragmentation algorithms for elastic optical networks using hitless spectrum retuning techniques , 2013, 2013 Optical Fiber Communication Conference and Exposition and the National Fiber Optic Engineers Conference (OFC/NFOEC).

[2]  P. J. Lin.  Reducing optical power variation in amplified optical network , 2003, International Conference on Communication Technology Proceedings, 2003. ICCT 2003..

[3]  L. Poti,et al.  Push-Pull Defragmentation Without Traffic Disruption in Flexible Grid Optical Networks , 2013, Journal of Lightwave Technology.

[4]  Gil Zussman,et al.  A machine learning approach for dynamic optical channel add/drop strategies that minimize EDFA power excursions , 2016 .

[5]  Houman Rastegarfar,et al.  Optical Power Dynamics in Wavelength Layer Software Defined Networking , 2015 .

[6]  Masahiko Jinno,et al.  Disruption minimized spectrum defragmentation in elastic optical path networks that adopt distance adaptive modulation , 2011, 2011 37th European Conference and Exhibition on Optical Communication.

[7]  Eiji Oki,et al.  Route partitioning scheme for elastic optical networks with hitless defragmentation , 2016, IEEE/OSA Journal of Optical Communications and Networking.

[8]  Daniel C. Kilper,et al.  Excursion-free dynamic wavelength switching in amplified optical networks , 2015, IEEE/OSA Journal of Optical Communications and Networking.