US 11,899,444 B2
Real-time alerts and transmission of selected signal samples under a dynamic capacity limitation
Ron Maurer, Haifa (IL); Sagi Schein, Haifa (IL); and Renato Keshet, Haifa (IL)
Assigned to Hewlett Packard Enterprise Development LP, Spring, TX (US)
Filed by Hewlett Packard Enterprise Development LP, Houston, TX (US)
Filed on Jun. 15, 2021, as Appl. No. 17/348,380.
Application 17/348,380 is a continuation of application No. 16/071,959, granted, now 11,061,395, previously published as PCT/US2016/016511, filed on Feb. 4, 2016.
Prior Publication US 2021/0311468 A1, Oct. 7, 2021
Int. Cl. G05B 23/02 (2006.01); G06N 3/044 (2023.01); H04L 9/40 (2022.01)
CPC G05B 23/0297 (2013.01) [G06N 3/044 (2023.01); H04L 63/1425 (2013.01)] 20 Claims
OG exemplary drawing
 
1. An alerting device of a production station, comprising:
processing circuitry; and
a memory including instructions that, when executed on the processing circuitry, cause the alerting device to:
receive a sample stream from a sensor of a set of sensors of the production station;
analyze the sample stream for anomalies in real time by inputting samples of the sample stream into an anomaly machine learning model trained to detect certain event patterns as anomaly patterns;
select a subset of the samples of the sample stream based on an extent of the anomaly patterns of the samples compared to a statistical model, wherein selecting the subset of samples includes determining a link capacity of a link connecting the production station to a base facility containing the statistical model;
transmit the selected subset of samples to the base facility;
receive an updated anomaly machine learning model, wherein the updated anomaly machine learning model is generated based, in part, on the transmitted subset of samples; and
upon receiving the updated anomaly machine learning model, analyze the sample stream for anomalies in real time by inputting samples of the sample stream into the updated anomaly machine learning model.