Photonic Reservoir Computing in Optical Transmission Systems

Hardware-based reservoir computing offers an alternative approach to conventional digital signal processing. Here two different types of photonic reservoir computers are evaluated with respect to their equalization performance in optical transmission systems. Feasibility and problems are also analyzed and possible solutions are discussed.

[1]  Marco Ruffini,et al.  An Overview on Application of Machine Learning Techniques in Optical Networks , 2018, IEEE Communications Surveys & Tutorials.

[2]  Simon Ohlendorf,et al.  100 km 56 GBd PAM-4 Transmission using Photonic Reservoir Computing , 2019, 45th European Conference on Optical Communication (ECOC 2019).

[3]  Herbert Jaeger,et al.  Reservoir computing approaches to recurrent neural network training , 2009, Comput. Sci. Rev..

[4]  S. Pachnicke,et al.  Photonic Reservoir Computing Enabled by Silicon Micro-Rings , 2019, 2019 Asia Communications and Photonics Conference (ACP).

[5]  Peter J. Winzer,et al.  From Scaling Disparities to Integrated Parallelism: A Decathlon for a Decade , 2017, Journal of Lightwave Technology.

[6]  Paul R. Prucnal,et al.  Photonic Reservoir Computing , 2017 .