Machine Learning-Based Multifunctional Optical Spectrum Analysis Technique

Wang, Danshi; Zhang, Min; Zhang, Zhiguo; Li, Jin; Gao, Hui; Zhang, Fan; Xue, Chengqi

doi:10.1109/access.2019.2895409

Cited by 25 publications

(19 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Optical spectrum analysis is required for measuring and monitoring the various characteristics of any optical fiber communication system. Currently, machine learning is used for performance enhancement of optical spectrum analysis technique [9]. It is impractical to design various optical communication systems in a laboratory.…”

Section: Need For Creation Of Fiber Optic Simulation Laboratorymentioning

confidence: 99%

Creation of an engineering course: design and simulation of high-capacity fiber optic systems utilizing VPI-photonics

Shadaram

Sarkar

Eghbal

2019

Fifteenth Conference on Education and Training in Optics and Photonics: ETOP 2019

View full text Add to dashboard Cite

In recent years, the importance of training skillful fiber optic engineers capable of designing and analyzing fiber optic links to deliver the enormous capacity needed for the next generation of mobile phones (5G) has been proven in industry. A comprehensive curriculum for such courses should undoubtedly comprise both theoretical and hands-on skills to better prepare the engineers before entering industry. The hands-on skills can be established through experimenting with real fiber optic links. However, having access to an equipped fiber optic laboratory imposes a significant cost to the academic institutions. Therefore, similar hands-on experiences can be delivered by software simulation of the physical signal generation and propagation. Thus, identifying an efficient yet powerful tool in simulating fiber optics components and devices is of utmost importance toward successfully rendering an understanding of the physical phenomena to the graduates. This paper aims to analyze the necessity of incorporating hands-on experience through software simulation and then it suggests the course topics and the learning outcomes expected to be achieved from this course. In addition, it investigates the feedbacks obtained from graduate engineering students throughout an actual course offered on fiber optic communication and demonstrates the assessment of the strength and weaknesses of the method. Finally, it introduces a powerful software package capable of simulating a variety of optical systems and devices frequently used in a typical fiber optic link and presents the performance analysis of this software package in addressing primitive and complex problems arising in fiber optics.

show abstract

Section: Need For Creation Of Fiber Optic Simulation Laboratorymentioning

confidence: 99%

Creation of an engineering course: design and simulation of high-capacity fiber optic systems utilizing VPI-photonics

Shadaram

Sarkar

Eghbal

2019

Fifteenth Conference on Education and Training in Optics and Photonics: ETOP 2019

View full text Add to dashboard Cite

show abstract

Section: Introductionmentioning

confidence: 99%

“…In [9], the authors identified support vector machine (SVM) regression as the most promising machine learning (ML) approach for OSNR estimation. However, [9] assumed a wide-band signal and in the majority of their evaluations used spectra coming from simulations. In our previous work [10], we also examined the performance of ML based models for in-band OSNR estimation, in particular SVM regression and a Gaussian process regression (GPR) model.…”

Section: Introductionmentioning

confidence: 99%

Experimental Demonstration of a Machine Learning-Based in-band OSNR Estimator from Optical Spectra

Locatelli¹,

Christodoulopoulos²,

Fàbrega³

et al. 2020

2020 International Conference on Optical Network Design and Modeling (ONDM)

View full text Add to dashboard Cite

Channel spectral monitors are becoming a cost effective solution to improve the management, resiliency and efficiency of next generation optical transport networks. We experimentally demonstrate a technique based on machine learning (ML) for the in-band estimation of amplified spontaneous emission (ASE) noise and filter 3-dB bandwidth, using optical spectra acquired after the reconfigurable optical add/drop multiplexers (ROADMs) filters. We assess the performance of the proposed method, considering laser drift and filters bandwidth tightening scenarios, showing quite good estimation accuracy under such conditions.

show abstract

“…Machine learning (ML) has recently been adopted in several scientific fields and is also becoming attractive in optical T communications. In [13], the authors considered four common ML models, and in particular, support vector machine (SVM), artificial neural network (ANN), k-nearest neighbors (KNN) and decision tree, and identified SVM as the most promising approach for OSNR estimation. However in [13], most of the spectral data were generated with a simulation tool and only few with experiments.…”

Section: Introductionmentioning

confidence: 99%

“…In [13], the authors considered four common ML models, and in particular, support vector machine (SVM), artificial neural network (ANN), k-nearest neighbors (KNN) and decision tree, and identified SVM as the most promising approach for OSNR estimation. However in [13], most of the spectral data were generated with a simulation tool and only few with experiments. Moreover, they considered classification with 1 dB accuracy, which is rather coarse, depending on the use case at hand.…”

Section: Introductionmentioning

confidence: 99%

Machine Learning-Based in-Band OSNR Estimation From Optical Spectra

Locatelli

Christodoulopoulos

Moreolo

et al. 2019

IEEE Photon. Technol. Lett.

View full text Add to dashboard Cite

Measuring the optical signal to noise ratio (OSNR) at certain network points is essential for failure handling, for single connection but also global network optimization. Estimating OSNR is inherently difficult in dense wavelength routed networks, where connections accumulate noise over different paths and tight filters do not allow the observation of the noise level at signal sides. We propose an in-band OSNR estimation process, which relies on a machine learning (ML) method, in particular on Gaussian process (GP) or support vector machine (SVM) regression. We acquired high-resolution optical spectra, through an experimental setup, using a Brillouin optical spectrum analyzer (BOSA), on which we applied our method and obtained excellent estimation accuracy. We also verified the accuracy of this approach for various resolution scenarios. To further validate it, we generated spectral data for different configurations and resolutions through simulations. This second validation confirmed the estimation quality of the proposed approach.

show abstract

Machine Learning-Based Multifunctional Optical Spectrum Analysis Technique

Cited by 25 publications

References 33 publications

Creation of an engineering course: design and simulation of high-capacity fiber optic systems utilizing VPI-photonics

Creation of an engineering course: design and simulation of high-capacity fiber optic systems utilizing VPI-photonics

Experimental Demonstration of a Machine Learning-Based in-band OSNR Estimator from Optical Spectra

Machine Learning-Based in-Band OSNR Estimation From Optical Spectra

Contact Info

Product

Resources

About