Associating machine-learning and analytical models for quality of transmission estimation: combining the best of both worlds

By associating machine learning and an analytical model (i.e., the Gaussian noise model), we reduce uncertainties on the output power profile and the noise figure of each amplifier in an optical network. We leverage the signal-to-noise ratio (SNR) of all the light paths of an optical network, monitored in all the coherent receivers. The learning process is based on a gradient-descent algorithm where all the uncertain input parameters of the analytical model are iteratively modified from their estimated values to match with the SNR of light paths in a European optical network. The design margin is then reduced to 0.1 dB for new traffic demands.

[1]  Konstantinos Christodoulopoulos,et al.  Formulating QoT Estimation with Machine Learning , 2018, 2018 European Conference on Optical Communication (ECOC).

[2]  Andrea Carena,et al.  Experimental Demonstration of Arbitrary Raman Gain-Profile Designs using Machine Learning , 2020, 2020 Optical Fiber Communications Conference and Exhibition (OFC).

[3]  Amirhossein Ghazisaeidi,et al.  Design Optimisation of Power-Efficient Submarine Line through Machine Learning , 2020, 2020 Conference on Lasers and Electro-Optics (CLEO).

[4]  Jelena Pesic,et al.  Transfer Learning Using ANN for G-OSNR Estimation in WDM Network Topologies , 2020 .

[5]  Joao Pedro,et al.  Fast and High-Precision Optical Performance Evaluation for Cognitive Optical Networks , 2020, 2020 Optical Fiber Communications Conference and Exhibition (OFC).

[6]  Andrea Carena,et al.  Inverse System Design Using Machine Learning: The Raman Amplifier Case , 2020, Journal of Lightwave Technology.

[7]  Amirhossein Ghazisaeidi,et al.  Experimental Prediction and Design of Ultra-Wideband Raman Amplifiers using Neural Networks , 2020, 2020 Optical Fiber Communications Conference and Exhibition (OFC).

[8]  Paparao Palacharla,et al.  Demonstration of Continuous Improvement in Open Optical Network Design by QoT Prediction using Machine Learning , 2019, 2019 Optical Fiber Communications Conference and Exhibition (OFC).

[9]  Dimitra Simeonidou,et al.  ANN-Based Multi-Channel QoT-Prediction Over a 563.4-km Field-Trial Testbed , 2020, Journal of Lightwave Technology.

[10]  S. Bigo,et al.  Learning process for reducing uncertainties on network parameters and design margins , 2018, 2017 Optical Fiber Communications Conference and Exhibition (OFC).

[11]  Roberto Proietti,et al.  Experimental demonstration of machine-learning-aided QoT estimation in multi-domain elastic optical networks with alien wavelengths , 2018, IEEE/OSA Journal of Optical Communications and Networking.

[12]  Massimo Tornatore,et al.  Reducing probes for quality of transmission estimation in optical networks with active learning , 2019, IEEE/OSA Journal of Optical Communications and Networking.

[13]  Luca Barletta,et al.  Machine-learning method for quality of transmission prediction of unestablished lightpaths , 2018, IEEE/OSA Journal of Optical Communications and Networking.

[14]  Zhiping Jiang,et al.  Machine Learning-Based EDFA Gain Model , 2018, 2018 European Conference on Optical Communication (ECOC).

[15]  Yvan Pointurier,et al.  Accurate QoT Estimation by Means of a Reduction of EDFA Characteristics Uncertainties with Machine Learning , 2020, 2020 International Conference on Optical Network Design and Modeling (ONDM).

[16]  Luca Barletta,et al.  QoT estimation for unestablished lighpaths using machine learning , 2017, 2017 Optical Fiber Communications Conference and Exhibition (OFC).

[17]  Massimo Tornatore,et al.  Using Active Learning to Decrease Probes for QoT Estimation in Optical Networks , 2019, 2019 Optical Fiber Communications Conference and Exhibition (OFC).

[18]  Daniel C. Kilper,et al.  Hybrid Machine Learning EDFA Model , 2020, 2020 Optical Fiber Communications Conference and Exhibition (OFC).

[19]  Min Zhang,et al.  QoT estimation for Unestablished Lightpaths Using Artificial Neural Networks , 2018, 2018 Conference on Lasers and Electro-Optics (CLEO).

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

[21]  Yiping Wang,et al.  Robust, compact, and flexible neural model for a fiber Raman amplifier , 2006 .

[22]  Zuyuan He,et al.  Intelligent Gain Flattening of FMF Raman Amplification by Machine Learning Based Inverse Design , 2020, 2020 Optical Fiber Communications Conference and Exhibition (OFC).

[23]  Takeshi Hoshida,et al.  Data-analytics-based Optical Performance Monitoring Technique for Optical Transport Networks , 2018, 2018 Optical Fiber Communications Conference and Exposition (OFC).

[24]  Andrea Carena,et al.  An ultra-fast method for gain and noise prediction of Raman amplifiers , 2019, 45th European Conference on Optical Communication (ECOC 2019).

[25]  Andrea Carena,et al.  Load Aware Raman Gain Profile Prediction in Dynamic Multi-Band Optical Networks , 2020, 2020 Optical Fiber Communications Conference and Exhibition (OFC).

[26]  A. Carena,et al.  Machine Learning-Based Raman Amplifier Design , 2018, 2019 Optical Fiber Communications Conference and Exhibition (OFC).

[27]  N Sambo,et al.  Lightpath Establishment Assisted by Offline QoT Estimation in Transparent Optical Networks , 2010, IEEE/OSA Journal of Optical Communications and Networking.

[28]  Christine Tremblay,et al.  Quality of transmission estimation and performance prediction of lightpaths using machine learning , 2019, 45th European Conference on Optical Communication (ECOC 2019).

[29]  Hao Jiang,et al.  Optimal Design of Gain-Flattened Raman Fiber Amplifiers Using a Hybrid Approach Combining Randomized Neural Networks and Differential Evolution Algorithm , 2018, IEEE Photonics Journal.

[30]  P. Poggiolini,et al.  A Simple and Effective Closed-Form GN Model Correction Formula Accounting for Signal Non-Gaussian Distribution , 2014, Journal of Lightwave Technology.

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

[32]  Ricardo Martinez,et al.  Modeling EDFA Gain Ripple and Filter Penalties With Machine Learning for Accurate QoT Estimation , 2020, Journal of Lightwave Technology.

[33]  P. Poggiolini.  The GN Model of Non-Linear Propagation in Uncompensated Coherent Optical Systems , 2012, Journal of Lightwave Technology.

[34]  J.-L Auge.  Can we use flexible transponders to reduce margins? , 2013, 2013 Optical Fiber Communication Conference and Exposition and the National Fiber Optic Engineers Conference (OFC/NFOEC).

[35]  Emanuele Virgillito,et al.  Using machine learning in an open optical line system controller , 2020, IEEE/OSA Journal of Optical Communications and Networking.

[36]  Ioannis Tomkos,et al.  Machine Learning for QoT Estimation of Unseen Optical Network States , 2019, 2019 Optical Fiber Communications Conference and Exhibition (OFC).

[37]  K. Christodoulopoulos,et al.  Quality of transmission estimation in WDM and elastic optical networks accounting for space–spectrum dependencies , 2016, IEEE/OSA Journal of Optical Communications and Networking.

[38]  Joao Pedro,et al.  Machine learning models for estimating quality of transmission in DWDM networks , 2018, IEEE/OSA Journal of Optical Communications and Networking.

[39]  Yvan Pointurier,et al.  Learning process for reducing uncertainties on network parameters and design margins , 2018, IEEE/OSA Journal of Optical Communications and Networking.

[40]  Ippokratis Sartzetakis,et al.  Accurate quality of transmission estimation with machine learning , 2019, IEEE/OSA Journal of Optical Communications and Networking.

[41]  Thierry Zami,et al.  How Uncertainty on the Fiber Span Lengths Influences QoT Estimation using Machine Learning in WDM Networks , 2020, 2020 Optical Fiber Communications Conference and Exhibition (OFC).

[42]  Alan Pak Tao Lau,et al.  Experimental Comparisons between Machine Learning and Analytical Models for QoT Estimations in WDM Systems , 2020, 2020 Optical Fiber Communications Conference and Exhibition (OFC).

[43]  Thierry Zami,et al.  The Perks of Using Machine Learning for QoT Estimation with Uncertain Network Parameters , 2020 .

[44]  Mohammad E. Mousa-Pasandi,et al.  Neural Network Training for OSNR Estimation from Prototype to Product , 2020, 2020 Optical Fiber Communications Conference and Exhibition (OFC).

[45]  Daniel C. Kilper,et al.  ANN-Based Transfer Learning for QoT Prediction in Real-Time Mixed Line-Rate Systems , 2018, 2018 Optical Fiber Communications Conference and Exposition (OFC).