Abstract
Techniques are provided herein for using a long-term trend inflection for optics parameters (such as optics error count) to make event predictions. Different methods (e.g., machine learning) may be used to model the inflection. Short-term or instantaneous optics measurements, such as pre-FEC error count, are often a reflection of instantaneous bursts or drifts and are thus not a reliable indicator for pervasive and inherent connection quality issues. A longer term and continuous trend analysis can better predict connection degradation and failure.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Zhang, Randy and Binetti, Stefano, "OPTICS LINE QUALITY PREDICTION THROUGH A LONG-TERM TREND INFLECTION", Technical Disclosure Commons, (November 04, 2022)
https://www.tdcommons.org/dpubs_series/5430