Multi-Layer Network Optimization Efficiently Exploiting Real-Time Performance Monitoring

Recent trends in optical transport networks emphasize the need of adopting a self-driving network in order to operate autonomously and adaptively to the actual network environment. In this context, advances in real-time telemetry platforms allowing to continuously monitor the current performance of an optical channel can avoid the traditional provisioning with end-of-life (EoL) margins, which is characterized by imposing margins large enough to account for worst-case performance degradation profiles (e.g. due to ageing) during network operation. Reducing these margins when provisioning an optical channel, can enable to operate it at higher capacity (e.g. using a higher order modulation format), thereby decreasing capital expenditures (CapEx). However, effectively adopting a more aggressive provisioning strategy requires a sophisticated multi-layer framework to address the higher likelihood that traffic demands might have a longer time to live than that of the optical channel carrying them. This paper exploits the joint availability of real-time performance monitoring platforms and OTN switching technology in order to develop a complete framework to provision services with reduced margins while proactively minimizing the utilization of optical channels that are approaching the performance limit, preventing traffic disruption. The simulation results highlight the benefits of adopting the proposed framework regarding the overall network performance and line interface count.