US 12,407,579 B2
Systems and methods for utilizing machine learning models to determine radio access network antenna performance impact
Jason A. Birr, Lithia, FL (US); Deborah Lynn Liske, Durham, NC (US); Richard S. Delk, Irmo, SC (US); and Brian A. Ward, Fort Worth, TX (US)
Assigned to Verizon Patent and Licensing Inc., Basking Ridge, NJ (US)
Filed by Verizon Patent and Licensing Inc., Basking Ridge, NJ (US)
Filed on Mar. 9, 2023, as Appl. No. 18/181,391.
Prior Publication US 2024/0305536 A1, Sep. 12, 2024
Int. Cl. H04L 41/16 (2022.01); H04L 41/0677 (2022.01); H04W 24/02 (2009.01); H04W 24/04 (2009.01); H04W 24/10 (2009.01)
CPC H04L 41/16 (2013.01) [H04L 41/0677 (2013.01); H04W 24/02 (2013.01); H04W 24/04 (2013.01); H04W 24/10 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method, comprising:
receiving, by a device, uplink physical resource block (PRB) interference data from each radio port of a base station;
processing, by the device, the uplink PRB interference data, with a first machine learning model, to generate same sector similarity score data for each radio port of the base station;
processing, by the device, the same sector similarity score data, with a second machine learning model and at predefined intervals, to identify and classify at least one antenna issue associated with the base station;
creating, by the device, sector-carrier pair data, from the same sector similarity score data and for a third machine learning model, based on the at least one antenna issue;
processing, by the device, the sector-carrier pair data, with the third machine learning model, to identify one or more issues that span sector carriers of the base station;
calculating, by the device, an issue status and alignment score based on the one or more issues that span the sector carriers of the base station and based on the same sector similarity score data; and
performing, by the device, one or more actions based on the issue status and alignment score.