US 11,949,531 B2
Systems and methods for proactive network diagnosis
Shane Yates, Atlanta, GA (US); Joseph Keller, Atlanta, GA (US); Christian Welin, Atlanta, GA (US); Sam Plant, Atlanta, GA (US); Tuba Aktaran-Kalayci, Atlanta, GA (US); Drew Milley, Atlanta, GA (US); Brian Stublen, Atlanta, GA (US); Alexis Hwang, Atlanta, GA (US); Mohammad Khabbazian, Atlanta, GA (US); and Will Gao, Atlanta, GA (US)
Assigned to Cox Communications, Inc., Atlanta, GA (US)
Filed by Cox Communications, Inc., Atlanta, GA (US)
Filed on Dec. 4, 2020, as Appl. No. 17/112,808.
Prior Publication US 2022/0182254 A1, Jun. 9, 2022
Int. Cl. H04L 12/28 (2006.01); G06N 20/00 (2019.01); H04N 7/173 (2011.01); H04N 21/4425 (2011.01); H04N 21/61 (2011.01)
CPC H04L 12/2801 (2013.01) [G06N 20/00 (2019.01); H04L 12/2856 (2013.01); H04N 7/17309 (2013.01); H04N 21/4425 (2013.01); H04N 21/6118 (2013.01)] 15 Claims
OG exemplary drawing
 
1. A system comprising:
at least one processor; and
at least one memory storing computer-executable instructions, that when executed by the at least one processor, cause the at least one processor to:
determine, at a first time, first telemetry data associated with a first subset of a group of cable modem devices, wherein the group of cable modem devices is associated with a cable network including one or more devices located downstream from a first network node device;
determine, at a second time, second telemetry data associated with a second subset of the group of cable modem devices, wherein the second subset of the group of cable modem devices is different than the first subset of the group of cable modem devices;
determine first streaming trap data indicative of an impairment of a first cable modem device of the second subset of the group of cable modem devices the, wherein the first telemetry data the first streaming trap data, and the second telemetry data are received from a cable modem termination system that communicates with the group of cable modem devices via the first network node device, wherein the first streaming trap data includes a micro-reflection;
determine, based on the second telemetry data indicating the impairment of the first cable modem device, a full poll of the one or more devices included within the cable network downstream from the first network node device, the full poll comprising:
identify a first network tap upstream from a first cable modem device of the group of cable modem devices;
determine third telemetry data associated with the first network tap;
identify all remaining network taps in the cable network; and
determine, subsequent to identifying the remaining network taps, fourth telemetry data associated with the remaining network taps and any cable modem devices located downstream from the remaining network taps;
generate first performance data associated with the first network node device;
determine, based on a comparison between the first performance data and an event criterion, an occurrence of an intermittent event associated with the first network node device;
determine, based on a machine learning model and the occurrence of only the intermittent event and without identifying a prior sustained event, a probability value indicative of a probability of a future occurrence of a sustained event;
determine that the probability value satisfies both a first threshold and a second threshold that is different than the first threshold; and
determine, based on the probability value satisfying the first threshold and second threshold, that a future sustained event will occur.