QoT estimation using EGN-assisted machine learning for multi-period network planning

The rapidly growing traffic demands in fiber-optical networks require flexibility and accuracy configuring lightpaths, for which fast accurate quality of transmission (QoT) estimation is pivotal importance. This paper introduces a machine learning (ML)-based QoT approach that meets these requirements. proposed gradient-boosting ML model uses precomputed per-channel self-channel-interference values as representative condensed features to estimate non-linear interference flexible-grid network. With an enhanced Gaussian noise (GN) simulation the baseline, achieves mean absolute signal-to-noise ratio error approximately 0.1 dB, improvement over GN model. For three different network topologies planning approaches varying complexities, multi-period study performed are compared path computation elements (PCEs). results show PCE capable matching or slightly improving performance on all while reducing significantly time by up 70%.