Artificial neural network based link OSNR estimation with a network approach

The performance monitoring of fiber-optics communication is an important task in nowadays communication system. Link optical noise-to-signal ratio (OSNR) is one of the most important parameters that affect the performance of optical networks. The traditional internal measurement method may increase the network construction cost and operation complexity. To overcome these drawbacks, an ANN based link OSNR estimation method with external measurement is proposed in this paper. Route level OSNR values are measured at the edge nodes and are used for link level OSNR estimation with the trained ANN. Besides, a heuristic method for route set generation is proposed to generate the route set that introduce fewer extra network load. The experiment results demonstrate that the ANN based method can meet the practical requirement in both estimation accuracy and computation complexity. The proposed method can be an important part of optical network OSNR monitoring to ensure robust and intelligent network operation.

[1]  Chao Lu,et al.  OSNR Monitoring for RZ-DQPSK Systems Using Half-Symbol Delay-Tap Sampling Technique , 2010, IEEE Photonics Technology Letters.

[2]  Chao Lu,et al.  Optical Performance Monitoring: A Review of Current and Future Technologies , 2016, Journal of Lightwave Technology.

[3]  J. Comellas,et al.  Performance monitor for all-optical networks based on homodyne spectroscopy , 2000, IEEE Photonics Technology Letters.

[4]  Deming Liu,et al.  Reference Optical Spectrum Based In-Band OSNR Monitoring Method for EDFA Amplified Multispan Optical Fiber Transmission System With Cascaded Filtering Effect , 2018, IEEE Photonics Journal.

[5]  Changyuan Yu,et al.  Optical signal to noise ratio monitoring using single channel sampling technique. , 2014, Optics express.

[6]  Chao Lu,et al.  Statistical Analysis of Optical Signal-to-Noise Ratio Monitoring Using Delay-Tap Sampling , 2010, IEEE Photonics Technology Letters.

[7]  A.E. Willner,et al.  Optical Performance Monitoring Using Artificial Neural Networks Trained With Eye-Diagram Parameters , 2009, IEEE Photonics Technology Letters.

[8]  A. Teixeira,et al.  Optical Signal-to-Noise Ratio Estimation Using Reference Asynchronous Histograms , 2009, Journal of Lightwave Technology.

[9]  Min Zhang,et al.  Convolutional Neural Network-Based Deep Learning for Intelligent OSNR Estimation on Eye Diagrams , 2017, 2017 European Conference on Optical Communication (ECOC).

[10]  Takeshi Hoshida,et al.  OSNR monitoring by deep neural networks trained with asynchronously sampled data , 2016, 2016 21st OptoElectronics and Communications Conference (OECC) held jointly with 2016 International Conference on Photonics in Switching (PS).

[11]  J.C. Li,et al.  Multi Impairment Monitoring for Optical Networks , 2009, Journal of Lightwave Technology.

[12]  J.H. Lee,et al.  A Review of the Polarization-Nulling Technique for Monitoring Optical-Signal-to-Noise Ratio in Dynamic WDM Networks , 2006, Journal of Lightwave Technology.