Abstract
A lot of algorithms are built to correlate two physical signals, the correlation output will be a mapping between the two signals data points. However, the correlation algorithms do not produce consistent quality along the two signals. Some regions might be correlated well while the others might be correlated poorly due to many effects. Our method computes a confidence score per each data point in the correlation mapping between the two signals. the score is calculated by shared sliding a window of defined size N thought the correlation mapping and apply linear regression on the two windows and reporting the R-squared metric of the window.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Baker Hughes Company, "A method to calculate the score of log correlation using sliding window linear regression", Technical Disclosure Commons, (April 07, 2025)
https://www.tdcommons.org/dpubs_series/7977