Dynamic data-driven power scaling in an optical node and network

A non-transitory computer-readable medium comprising instructions that, when executed, cause a processor to perform the steps of:

• obtaining power measurements from one or more first components in an optical node and store the power measurements at corresponding vertices in a graph that describes the optical node, wherein the power measurements include at least two power measurements with a first measurement and a second measurement;
• obtaining calibration data from a plurality of second components in the optical node and utilize the calibration data to determine an optical gain or an optical loss through an associated arc in the graph;
• determining an optical power at any point in the optical node via traversing the graph, for any of configuration and operation of the optical node, wherein the optical power is determined at the point by traversing the graph from the first measurement to the point and traversing the graph from the second measurement to the point;
• and
• determine an indication of improper operation of the node based on a divergence of the determined optical power based on the first measurement and the second measurement.

A non-transitory computer-readable medium comprising instructions that, when executed, cause a processor to perform the steps of:

• obtaining power measurements from one or more first components in an optical node and store the power measurements at corresponding vertices in a graph that describes the optical node;
• obtaining calibration data from a plurality of second components in the optical node and utilize the calibration data to determine an optical gain or an optical loss through an associated arc in the graph;
• and
• determining an optical power at any point in the optical node via traversing the graph, for any of configuration and operation of the optical node,
• wherein an amplifier component of the second components is modeled by an arc with an optical gain, a fiber connection of the second components is modeled by an arc with an optical loss based on fiber attenuation, and a circuit pack of the second components is modeled by an arc with an optical loss based on insertion loss.