Supervised graph convolution networks for OSNR and power estimation in optical mesh networks
Bibliographic record
Abstract
The optical signal-to-noise ratio (OSNR) and received optical channel power are critical parameters in determining the quality of transmission. The OSNR and received optical channel power are influenced by network impairments such as fiber loss, amplified stimulated emission noise, and nonlinear impairments. Furthermore, environmental effects and routing, modulation, and spectrum assignment schemes influence the OSNR and thus the reach of the optical channels. These impairments and effects vary with the spectral loads that are hard to predict in brownfield networks. Several deep neural network (DNN)-based methods have been explored to estimate the OSNR and nonlinear noise. However, these methods ignore the network topology. This paper bridges this gap by leveraging supervised graph convolution neural networks (GCNs), which operate directly on graphs for OSNR and received power estimation in an optical mesh network. We also develop and implement a novel graph windowed neural network (GWinN) to reduce the over-smoothing effects of a GCN and thus learn localized behaviors like fiber cuts. We apply a DNN, GCN, and GWinN in practice to a testbed of 8 reconfigurable optical add-drop multiplexers and 22 amplifiers. Our procedure accurately estimates the OSNR with a prediction error mean and a standard deviation of ( − 0.02 d B , 0.35 dB) for a reference OSNR ranging from (16 dB) to (24 dB).
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.004 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".