公开(公告)号：US11861453B2

公开(公告)日：2024-01-02

申请号：US17979508

申请日：2022-11-02

Applicant: GOOGLE LLC

Inventor： James Peroulas ,  Poojita Thukral ,  Dutt Kalapatapu ,  Andreas Terzis ,  Krishna Sayana

IPC: G06N3/08 ,  H04W24/08 ,  H04W84/04

CPC classification number: G06N3/08 ,  H04W24/08 ,  H04W84/042

Abstract: A method for detecting network anomalies includes receiving a control message from a cellular network and extracting one or more features from the control message. The method also includes predicting a potential label for the control message using a predictive model configured to receive the one or more extracted features from the control message as feature inputs. Here, the predictive model is trained on a set of training control messages where each training control message includes one or more corresponding features and an actual label. The method further includes determining that a probability of the potential label satisfies a confidence threshold. The method also includes analyzing the control message to determine whether the control message corresponds to a respective network performance issue. When the control message impacts network performance, the method includes communicating the network performance issue to a network entity responsible for the network performance issue.

公开(公告)号：US20220345900A1

公开(公告)日：2022-10-27

申请号：US17811597

申请日：2022-07-11

Applicant: Google LLC

Inventor： James Peroulas ,  Sharath Ananth ,  Nachiappan Valliappan

IPC: H04W16/18 ,  H04W24/06 ,  H04W88/08 ,  H04W48/16 ,  H04W16/00

Abstract: A method for detecting coverage problems is provided. The method includes receiving, at data processing hardware, from at least one user equipment (UE), observations. Each observation includes a signal measurement of a signal emitted from a base station and a corresponding location of the signal measurement. The method also includes generating, by the data processing hardware, a coverage map for the base station based on the received observations, the coverage map indicating a signal characteristic of the emitted signal about the base station. The method further includes determining, by the data processing hardware, an estimated characteristic of the base station by feeding the coverage map into a neural network configured to output the estimated characteristic of the base station.

公开(公告)号：US20210133574A1

公开(公告)日：2021-05-06

申请号：US17145236

申请日：2021-01-08

Applicant: Google LLC

Inventor： James Peroulas ,  Poojita Thukral ,  Dutt Kalapatapu ,  Andreas Terzis ,  Krishna Sayana

IPC: G06N3/08 ,  H04W24/08

Abstract: A method for detecting network anomalies includes receiving a control message from a cellular network and extracting one or more features from the control message. The method also includes predicting a potential label for the control message using a predictive model configured to receive the one or more extracted features from the control message as feature inputs. Here, the predictive model is trained on a set of training control messages where each training control message includes one or more corresponding features and an actual label. The method further includes determining that a probability of the potential label satisfies a confidence threshold. The method also includes analyzing the control message to determine whether the control message corresponds to a respective network performance issue. When the control message impacts network performance, the method includes communicating the network performance issue to a network entity responsible for the network performance issue.

公开(公告)号：US10891546B2

公开(公告)日：2021-01-12

申请号：US16397082

申请日：2019-04-29

Applicant: Google LLC

Inventor： James Peroulas ,  Poojita Thukral ,  Dutt Kalapatapu ,  Andreas Terzis ,  Krishna Sayana

IPC: G06N3/08 ,  H04W24/08 ,  H04W84/04

Abstract: A method for detecting network anomalies includes receiving a control message from a cellular network and extracting one or more features from the control message. The method also includes predicting a potential label for the control message using a predictive model configured to receive the one or more extracted features from the control message as feature inputs. Here, the predictive model is trained on a set of training control messages where each training control message includes one or more corresponding features and an actual label. The method further includes determining that a probability of the potential label satisfies a confidence threshold. The method also includes analyzing the control message to determine whether the control message corresponds to a respective network performance issue. When the control message impacts network performance, the method includes communicating the network performance issue to a network entity responsible for the network performance issue.

公开(公告)号：US20190268779A1

公开(公告)日：2019-08-29

申请号：US15903756

申请日：2018-02-23

Applicant: Google LLC

Inventor： James Peroulas ,  Sharath Ananth ,  Nachiappan Valliappan

IPC: H04W16/18 ,  H04W24/06 ,  H04W88/08 ,  H04W48/16

Abstract: A method for detecting coverage problems is provided. The method includes receiving, at data processing hardware, from at least one user equipment (UE), observations. Each observation includes a signal measurement of a signal emitted from a base station and a corresponding location of the signal measurement. The method also includes generating, by the data processing hardware, a coverage map for the base station based on the received observations, the coverage map indicating a signal characteristic of the emitted signal about the base station. The method further includes determining, by the data processing hardware, an estimated characteristic of the base station by feeding the coverage map into a neural network configured to output the estimated characteristic of the base station.