Quality of Transmission Estimation in WDM and Elastic Optical Networks Accounting for Space–Spectrum Dependencies

Sartzetakis, Ippokratis; Christodoulopoulos, K.; Tsekrekos, Christos P.; Syvridis, Dimitris; Varvarigos, Emmanouel

doi:10.1364/jocn.8.000676

Cited by 49 publications

(15 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Several heuristic algorithms for the placement of the probes are proposed in [34]. A further refinement of the methodologies which takes into account the presence of neighbor channels appears in [35].…”

Section: A Quality Of Transmission Estimationmentioning

confidence: 99%

An Overview on Application of Machine Learning Techniques in Optical Networks

Musumeci

Rottondi

Nag

et al. 2019

IEEE Commun. Surv. Tutorials

486

226

View full text Add to dashboard Cite

Today's telecommunication networks have become sources of enormous amounts of widely heterogeneous data. This information can be retrieved from network traffic traces, network alarms, signal quality indicators, users' behavioral data, etc. Advanced mathematical tools are required to extract meaningful information from these data and take decisions pertaining to the proper functioning of the networks from the network-generated data. Among these mathematical tools, Machine Learning (ML) is regarded as one of the most promising methodological approaches to perform network-data analysis and enable automated network self-configuration and fault management.The adoption of ML techniques in the field of optical communication networks is motivated by the unprecedented growth of network complexity faced by optical networks in the last few years. Such complexity increase is due to the introduction of a huge number of adjustable and interdependent system parameters (e.g., routing configurations, modulation format, symbol rate, coding schemes, etc.) that are enabled by the usage of coherent transmission/reception technologies, advanced digital signal processing and compensation of nonlinear effects in optical fiber propagation.In this paper we provide an overview of the application of ML to optical communications and networking. We classify and survey relevant literature dealing with the topic, and we also provide an introductory tutorial on ML for researchers and practitioners interested in this field. Although a good number of research papers have recently appeared, the application of ML to optical networks is still in its infancy: to stimulate further work in this area, we conclude the paper proposing new possible research directions.

show abstract

Section: A Quality Of Transmission Estimationmentioning

confidence: 99%

An Overview on Application of Machine Learning Techniques in Optical Networks

Musumeci

Rottondi

Nag

et al. 2019

IEEE Commun. Surv. Tutorials

486

226

View full text Add to dashboard Cite

show abstract

“…A similar approach is adopted in [12] to tune design margins in presence of unknown network parameters: a controller collects and stores field data, which are ingested by a prediction tool that outputs an OSNR estimation based on educated guesses on the unknown network parameters. In [13], Gaussian-Processes nonlinear Regression is adopted to predict the BER of an optical communication system using as features the channel input power, lightpath length, symbol rate and inter-channel spacing.Two alternative ML approaches named network kriging and norm L 2 minimization are applied in [14]- [17] to perform QoT estimation. These methods require the installation of probe lightpaths carrying dummy traffic to acquire field measurements to compute an estimation of the Q-factor of already established or candidate lightpaths.…”

Section: A Quality-of-transmission Estimationmentioning

confidence: 99%

Routing and Spectrum Assignment Integrating Machine-Learning-Based QoT Estimation in Elastic Optical Networks

Salani

Rottondi

Tornatore

2019

IEEE INFOCOM 2019 - IEEE Conference on Computer Communications

View full text Add to dashboard Cite

Machine Learning (ML) is under intense investigation in optical networks as it promises to lead to automation of a variety of management tasks, as amplifier gain equalization, fault recognition, Quality of Transmission (QoT) estimation, and many others. Though several studies focus on each of these specific tasks, the integration of ML-based estimations inside Routing and Spectrum Assignment (RSA) is still largely unexplored. This paper moves towards such integration. We develop a framework that leverages the probabilistic outputs of a MLbased QoT estimator to define the reach constraints in an Integer Linear Programming (ILP) formulation for RSA in an elastic optical network. In this integrated procedure, the RSA problem is solved iteratively by updating the reach constraints based on the outcome of a QoT estimator, to exclude lightpaths with unacceptable QoT. In our numerical evaluation, the proposed integrated method achieves savings in spectrum occupation up to 30% (around 20% on average) compared to traditional ILPbased RSA approaches with reach constraints based on margined analytical models.

show abstract

“…Research contributions of the first type assume that each channel in the network adheres to an signal-to-noise ratio (SNR) threshold. Thereafter, spectrum allocation, PSD, modulation format, and routing is decided for the channels as a function of the traffic demand [7], [8], [11]- [13], whereas research contributions of the second type observe the effect of provisioning (activating) new light paths on the QoT of all channels [14].…”

Section: Introductionmentioning

confidence: 99%

Channel allocation in elastic optical networks using traveling salesman problem algorithms

Bhar

Agrell

Keykhosravi

et al. 2019

J. Opt. Commun. Netw.

View full text Add to dashboard Cite

Elastic optical networks have been proposed to support high data-rates in metro and core networks. However, frequency allocation of the channels (i.e., channel ordering) in such networks is a challenging problem. This requires arrangement of the optical channels within the frequency grid with the objective of ensuring a minimum signal-to-noise ratio (SNR). An optimal arrangement results in the highest SNR margin for the entire network. However, searching for the optimal arrangement requires an exhaustive search through all possible arrangements (permutations) of the channels. The search space increases exponentially with the number of channels. This discourages an algorithm employing exhaustive search for the optimal frequency allocation. We utilize the Gaussian noise (GN) model to formulate the frequency allocation (channel ordering) problem as a variant of the traveling salesman problem (TSP) using graph theory. Thereafter, we utilize graph-theoretic tools for TSP from the existing literature to solve the channel ordering problem. Performance figures obtained for the proposed scheme is illustrated to be marginally inferior to the optimal search (through all possible permutations) and outperforms any random allocation scheme. Moreover, the proposed scheme is implementable for a scenario with a large number of channels. In comparison, exhaustive search with the GN model and splitstep Fourier method simulations are shown to be feasible for a small number of channels only. It is also illustrated that the SNR reduces with an increase in bandwidth when the frequency separation is high.

show abstract

Quality of Transmission Estimation in WDM and Elastic Optical Networks Accounting for Space–Spectrum Dependencies

Cited by 49 publications

References 9 publications

An Overview on Application of Machine Learning Techniques in Optical Networks

An Overview on Application of Machine Learning Techniques in Optical Networks

Routing and Spectrum Assignment Integrating Machine-Learning-Based QoT Estimation in Elastic Optical Networks

Channel allocation in elastic optical networks using traveling salesman problem algorithms

Contact Info

Product

Resources

About