Revisiting the calculation of performance margins in monitoring-enabled optical networks

Receiver-Based Experimental Estimation of Power Losses in Optical Networks

“…Moreover, to describe the time-varying network conditions, a polarization state monitoring method was proposed in [4]. In [5], it has been shown that the power attenuation uncertainty has a significant impact on the system characterization. Therefore, accurately estimating power losses -which may come from abnormal splicing, excessing connector loss, intrusion, or tapping-enables us to make the best possible decision in terms of operational cost or outage avoidance.…”

Optical path management based on machine learning for optical networks

“…The signal quality required during a service and the degree of uncertainty determine the margin needed. Margins can be classified into system margins, design margins, and unallocated margins as summarized in Table 1 [5,6,12,13]. The system margin is provided to handle time-varying deterioration factors.…”

Flexible Technologies to Increase Optical Network Capacity

“…These margins typically account for uncertainties both in the amount of traffic to be supported, but also in the knowledge of some of the physical parameters of the fibre plant, that might differ from the initial design value and/or might change over time due to the aging of equipment [34]. One way to reduce these margins is to collect detailed monitoring data [35] (monitors can be deployed at receivers, as well as at ROADMs, in the form of spectrum analyzers), then using Machine Learning (ML) to estimate relevant parameters [36] whose current knowledge is otherwise inaccurate.…”