Protection against failure of machine-learning-based QoT prediction

Abstract

Machine learning (ML)-based methods are being widely explored to predict the quality of transmission (QoT) of a lightpath. They are expected to reduce the signal-to-noise ratio margin reserved for the lightpath, thus improving the spectrum efficiency of an optical network. However, many studies on this prediction are often based on synthetic datasets or datasets obtained from laboratories. As such, these datasets may not accurately represent the entire state space of a practical optical network, which is exposed in harsh environments. There are risks of failure when using these ML-based QoT prediction models. Thus, it is necessary to develop a mechanism that can guarantee the reliability of lightpath service even if the prediction model fails. For this scenario, we propose to employ the conventional network protection techniques that are usually implemented to protect against network node/link failures to also protect against the failure of QoT prediction. Based on the two representative protection techniques, i.e., 1 + 1 dedicated path protection and shared backup path protection (SBPP), the performance of the proposed protection mechanism is evaluated by reserving different margins for the working and protection lightpaths. For 1 + 1 path protection, we find that the proposed mechanism can achieve a zero design margin (D-margin) for a working lightpath, thereby significantly improving network spectrum efficiency, while not sacrificing the availability of lightpath services. For SBPP, we find that an optimal D-margin should be identified to balance the spectrum efficiency and service availability; the proposed mechanism can save up to 0.5-dB D-margin for a working lightpath, while guaranteeing service availability.

ML-assisted restoration planning and upgrade with low design margins

“…However, in real-life these inputs are often not known precisely 5 , and safety design margins are imposed to guarantee that modulation format assigned to the lightpath based on predicted QoT is feasible in the field deployment. The extent of these margins depends on the available information about the network and its size, but can easily reach 2-3 dB in core networks 6 , leading to significant under-utilization of resources. Notable research effort has been recently dedicated to lowering these margins by either estimating the precise values of uncertain input parameters 7, 8, 9 or directly predicting QoT metrics using measurements from previously established lightpaths 10 .…”

“…The only existing work that combines ML-based low-margin design and resilience in optical networks is 6 , where authors demonstrate savings from ML-based QoT-estimation for dedicated and shared protection. In this work, for the first time to the best of our knowledge, we investigate possible savings from ML-estimated design margins in 2 restoration scenarios: 1) Restoration Planning and 2) Restoration Upgrade.…”

Probabilistic low-margin optical-network design with multiple physical-layer parameter uncertainties

“…However, in reallife, PL parameters such as the EDFA gain ripple, connector loss and, sometimes, even the fiber type are not known precisely, so safety (design) margins [3] are imposed to guarantee that the MF configured based on predicted QoT is operable in the field deployment, and the correct number of transponders is installed to satisfy traffic requests. The extent of design margins depends on the available information about the network, its size and available monitoring capabilities, but, to account for the worst-case deviation of the predicted vs. the actual SNR, in presence of multiple PL uncertainties, they can easily reach up to (2)(3) dB in core networks [4], leading to significant resource under-utilization.…”

Machine learning-based QoT estimation over optical wireless communication

“…Currently, humanity demands the application of a communication system by using machine learning algorithms. The regression algorithms are more flexible, and reliable [3]. Regression algorithms can improve the accuracy and performance of fiber-optic communication and networks.…”