We propose a parameter refinement method based on incremental learning, leveraging multiple network snapshots to provide accurate estimated inputs (i.e., lumped losses, gain spectra, and offset noise) to digital twins, improving QoT prediction and optimization.
Inputs Refinement with Incremental Learning for Accurate Digital Twin of Optical Networks
X. Yang;M. Tornatore;
2025-01-01
Abstract
We propose a parameter refinement method based on incremental learning, leveraging multiple network snapshots to provide accurate estimated inputs (i.e., lumped losses, gain spectra, and offset noise) to digital twins, improving QoT prediction and optimization.File in questo prodotto:
| File | Dimensione | Formato | |
|---|---|---|---|
|
Inputs_Refinement_with_Incremental_Learning_for_Accurate_Digital_Twin_of_Optical_Networks.pdf
accesso aperto
Dimensione
697.43 kB
Formato
Adobe PDF
|
697.43 kB | Adobe PDF | Visualizza/Apri |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
