论文信息 - What if AI Fails: Protection against Failure of AI-Based QoT Prediction

What if AI Fails: Protection against Failure of AI-Based QoT Prediction

We propose and simulate a new mechanism to protect against the failure of AI-based QoT prediction by assigning OSNR margins for working lightpath using the AI-based prediction method and for protection lightpath using the traditional, conservative method. This guarantees reliability of lightpath services, while not significantly increasing network spectrum resources used.

Gangxiang Shen | Sanjay K. Bose | Lian Xiang | Ningning Guo | Longfei Li

[1] P. Poggiolini,et al. The GN-Model of Fiber Non-Linear Propagation and its Applications , 2014, Journal of Lightwave Technology.

[2] Rafael Rios Müller,et al. Advanced modulation formats and signal processing for high speed spectrally efficient optical communications , 2016 .

[3] Yvan Pointurier,et al. Design of low-margin optical networks , 2017, IEEE/OSA Journal of Optical Communications and Networking.

[4] Mingyi Gao,et al. Migrating elastic optical networks from standard single-mode fibers to ultra-low loss fibers: Strategies and benefits , 2017, 2017 Optical Fiber Communications Conference and Exhibition (OFC).

[5] Massimo Tornatore,et al. Routing and Spectrum Assignment Integrating Machine-Learning-Based QoT Estimation in Elastic Optical Networks , 2019, IEEE INFOCOM 2019 - IEEE Conference on Computer Communications.

[6] Gangxiang Shen,et al. Impairment-Aware Lightpath Routing and Regenerator Placement in Optical Transport Networks With Physical-Layer Heterogeneity , 2011, Journal of Lightwave Technology.