公开(公告)号：US20250068546A1

公开(公告)日：2025-02-27

申请号：US18452599

申请日：2023-08-21

Applicant: HEWLETT PACKARD ENTERPRISE DEVELOPMENT LP

Inventor： Rahul Bahal ,  Derwin Thulani Ngomane ,  Laura Lee Richter

IPC: G06F11/36 ,  G01D21/00

Abstract: In some examples, different time offsets within a test time interval are computed for tests by respective sensors of a plurality of sensors, where a time offset of the different offsets indicates a point in time within the test time interval with respect to which a corresponding sensor is to initiate a test. A sensor of the plurality of sensors performs tests according to a corresponding time offset of the different time offsets.

公开(公告)号：US20240205100A1

公开(公告)日：2024-06-20

申请号：US18085380

申请日：2022-12-20

Applicant: HEWLETT PACKARD ENTERPRISE DEVELOPMENT LP

Inventor： LAURA RICHTER ,  Leonard Botha ,  Derwin Thulani Ngomane ,  Stephen Hulme ,  Njabulo Ndlovu

IPC: H04L41/147 ,  H04L41/06 ,  H04L41/08 ,  H04L41/14 ,  H04L41/40 ,  H04L43/16

CPC classification number: H04L41/147 ,  H04L41/06 ,  H04L41/0886 ,  H04L41/145 ,  H04L41/40 ,  H04L43/16

Abstract: Systems and methods are provided for automated anomaly detection model quality assurance (QA) and deployment for wireless network failure prediction. Network failure prediction can leverage models trained to detect issues on a network and predict failure scenarios by identifying anomalous issues indicative of failure conditions. To keep these models up to date with changes in network behavior and configurations, the models are recalibrated from time to time. Implementations disclosed herein provide for automated evaluation and deployment of recalibrated models, while assuring issue detection results from the recalibrated models accurately reflect current network conditions. To do this, implementations disclosed herein determine QA metrics for recalibrated, candidate models, QA thresholds from previously deployed models, and QA criteria from a currently deployed model. Based on a comparison of the QA metrics with the QA thresholds and the QA criteria, implementations disclosed herein automatically deploy recalibrated, candidate models without human or external intervention.