US 12,279,002 B2
Method and system for quantifying effects of a content delivery network server on streaming-media quality and predicting root cause analysis
Xiaowen Mang, Morganville, NJ (US); Carolyn Roche Johnson, Holmdel, NJ (US); Gregory J. Smith, Shrewsbury, NJ (US); and Peni-Taito Payne, Katy, TX (US)
Assigned to AT&T Intellectual Property I, L.P., Atlanta, GA (US)
Filed by AT&T Intellectual Property I, L.P., Atlanta, GA (US)
Filed on May 4, 2022, as Appl. No. 17/736,457.
Prior Publication US 2023/0362420 A1, Nov. 9, 2023
Int. Cl. H04N 21/24 (2011.01); H04N 21/442 (2011.01); H04N 21/466 (2011.01)
CPC H04N 21/2402 (2013.01) [H04N 21/2404 (2013.01); H04N 21/44204 (2013.01); H04N 21/4666 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method comprising:
calculating, by a processing system including a processor, a threshold value of a streaming-media key performance indicator (KPI) based on a level of certainty that a predetermined target portion of a plurality of end-user media processors provide an acceptable end-user experience;
obtaining, by the processing system, access records of a content delivery network (CDN) configured to serve media content requested by the plurality of end-user media processors, wherein the CDN includes at least one origin server and at least one streaming-content caching server;
predicting, by the processing system, a plurality of values of the streaming-media KPI according to the access records of the CDN, to obtain a plurality of predicted values of the streaming-media KPI;
comparing, by the processing system, the plurality of predicted values of the streaming-media KPI to the threshold value of the streaming-media KPI within a sequence of anomaly threshold time windows;
determining, by the processing system, that a threshold number of the plurality of predicted values of the streaming-media KPI exceed the threshold value of the streaming-media KPI in at least one of the anomaly threshold time windows indicating a need for capacity growth in the CDN; and
identifying, by the processing system, an anomaly according to the determining, the anomaly indicating that the need for capacity growth is in one of the at least one origin server and the at least one streaming-content caching server.