Multi-period traffic on elastic optical networks planning: alleviating the capacity crunch

Hybrid strategy for routing, modulation and spectrum assignment in elastic optical networks

“…After that, the network design is verified and manually implemented. While the network is in operation, its capacity can be continuously monitored and the resulting data is used as input for the next planning cycle (López et al, 2016;Mesquita et al, 2020). In case of unexpected increase in demand or network changes, nonetheless, the planning process may be restarted.…”

Uncertainty quantification and consideration in ML-aided traffic-driven service provisioning

“…Both supervised learning (SL) and reinforcement learning (RL) techniques have been leveraged in the state of the art for network (re)optimization and (re)configuration purposes [5]. Specifically, SL is used for training a predictive ML model given a dataset of past traffic demand traces with the traffic predictions used to (re)optimize the network resources (i.e., predictive service provisioning) [2,[8][9][10][11][12][13][14][15][16][17], while RL is used for learning (re)optimization policies through a trial-and-error process (i.e., prescriptive provisioning) [3,[18][19][20][21].…”

“…While this problem has been extensively studied in the literature, targeting to optimize various objectives, such as service disruptions [9][10][11], spectrum utilization [8,12], energy utilization [13][14][15], and QoSbased fairness [16], the uncertainty in the traffic predictions, possibly leading to erroneous decisions (i.e., violations on the QoS requirements), has been largely ignored [5]. Hence, in this work, the focus is on addressing uncertainty in traffic predictions, aiming to efficiently (re)optimize the network resources, while at the same time preserving targeted QoS requirements at acceptable levels.…”