The massive deployment of 5G and beyond will require high capacity and low latency connectivity services, so network operators will have either to overprovision capacity in their transport networks or to upgrade the optical network controllers to make decisions nearly in real time; both solutions entail high capital and operational expenditures. A different approach could be to move the decision making toward the nodes and subsystems, so they can adapt dynamically the capacity to the actual needs and thus reduce operational costs in terms of energy consumption. To achieve this, several technological challenges need to be addressed. In this paper, we focus on the autonomous operation of Digital Subcarrier Multiplexing (DSCM) systems, which enable the transmission of multiple and independent subcarriers (SC). Herein, we present several solutions enabling the autonomous DSCM operation, including: i) SC quality of transmission estimation; ii) autonomous SC operation at the transmitter side and blind SC configuration recognition at the receiver side; and iii) intent-based capacity management implemented through Reinforcement Learning. We provide useful guidelines for the application of autonomous SC management supported by the extensive results presented.

Within a context of software-defined optical transport networks (SD-OTN), this work addresses specifically the management of Raman amplification, aiming to introduce and experimentally validate a system able to autonomously control this feature in-situ. In particular, given the required amplification constraints, an ad-hoc software module has been developed in order to optimize Raman pump power levels. Then, relying on this software, the architecture of an embedded controller to install on board the Raman card has been defined to handle Raman pumps. The use of a conceived probing procedure allows to self-adapt each Raman amplifier to the installed fiber, allowing it to autonomously operate at the working point required by the control plane. Relying on the system telemetry, the proposed architecture controls the Raman pumps in order to achieve the required amplification constraints in terms of average gain and tilt. The entire proposal is validated through a set of experimental measurements that proofs both the achievement of the required gain target and the importance of the probing phase procedure in making the Raman amplifier autonomous and self-adaptable.

GNPy is an open source project of the Telecom Infra Project (TIP). It aims at the development of a software library for the abstraction of physical layer data transport, implemented as wavelength division multiplexing (WDM) through open-source code and open APIs, with the purpose of automatizing disaggregated, multi-vendor open optical networks. We first introduce the GNPy library of applications and functionalities, then, we present validation of the GNPy core. The quality-of-transmission estimator (QoT-E) enables a quick and accurate prediction of the unique performance metric in WDM data transport of optical networks: the Generalized Signal-to-Noise Ratio (GSNR). The presented validation experiments show an excellent accuracy in predicting the GSNR compared to the values obtained experimentally. Finally, we present the GNPy integration with the ONOS network OS in a multi-vendor scenario including open reconfigurable add drop multiplexers (ROADMs) and the related validation experiment.