Systems and methods for managing a wireless network based on user equipment data

Patent metadata

Field	Value
Publication number	US-12580807-B2
Application number	US-202318535000-A
Country	US
Kind code	B2
Filing date	Dec 11, 2023
Priority date	Dec 11, 2023
Publication date	Mar 17, 2026
Grant date	Mar 17, 2026

Abstract

In some implementations, a device may identify a network node that supports a first radio access technology (RAT), wherein the network node is associated with sessions with user equipments (UEs). The device may calculate a percentage of traffic from the sessions that is associated with the first RAT and not a second RAT. The device may determine that the network node is associated with a score based on the percentage of traffic associated with the first RAT not satisfying a threshold. The device may collect, based on the network node being associated with the score, device data associated with the UEs. The device may determine a network issue associated with the network node based on the device data. The device may determine a root cause for the network issue. The server may transmit an indication of the root cause.

First claim

What is claimed is:

1 . A method, comprising:
identifying, by a server, a set of user equipments (UEs) that are each connected to a network node for an amount of time that satisfies a first threshold, wherein the network node supports a first radio access technology (RAT); identifying, by the server and from the set of UEs, a subset of UEs associated with sessions, wherein the sessions each exchange an amount of data that satisfies a second threshold; calculating, by the server, a percentage of traffic from the sessions that is associated with the first RAT and not a second RAT; determining, by the server, that the network node is associated with a score based on the percentage of traffic associated with the first RAT not satisfying a third threshold; collecting, by the server and based on the network node being associated with the score, device data associated with the subset of UEs, wherein the device data is associated with signals corresponding to a cell addition procedure; determining, by the server, a network issue associated with the network node based on the device data; determining, via a machine learning (ML) model that runs on the server, a root cause for the network issue; and transmitting, by the server, an indication of the root cause.

2 . The method of claim 1 , wherein the root cause triggers an entity in a wireless network to perform an autonomous corrective action to resolve the network issue, and a type of autonomous corrective action is based on the root cause.

3 . The method of claim 1 , wherein the first RAT is a New Radio (NR) RAT and the second RAT is a Long Term Evolution (LTE) RAT.

4 . The method of claim 1 , further comprising:
verifying that a device model associated with a UE, of the subset of UEs, and a price plan associated with the UE support the first RAT.

5 . The method of claim 1 , wherein the signals corresponding to the cell addition procedure include signals associated with a first radio resource control (RRC) reconfiguration request with a measurement identifier and a band requirement.

6 . The method of claim 1 , wherein the signals corresponding to the cell addition procedure include signals associated with a measurement report from a UE, of the subset of UEs, for a measurement identifier.

7 . The method of claim 1 , wherein the signals corresponding to the cell addition procedure include signals associated with a second radio resource control (RRC) reconfiguration request with a physical cell identifier (PCI) and a absolute radio-frequency channel number (ARFCN).

8 . The method of claim 1 , wherein the signals corresponding to the cell addition procedure include signals associated with a successful random access channel (RACH) attach message.

9 . The method of claim 1 , wherein the signals corresponding to the cell addition procedure include signals associated with a connection release message.

10 . The method of claim 1 , wherein the network issue is associated with an abnormal pattern resulting from a plurality of sessions associated with the network node ending during the cell addition procedure.

11 . The method of claim 1 , wherein the network issue is associated with a frequent release of connections associated with the subset of UEs, and the frequent release occurs when the connections are released within an amount of time after a time of connection establishment that satisfies a fourth threshold.

12 . The method of claim 1 , wherein determining, via the ML model, the root cause for the network issue is based on a comparison of features associated with the network issue and features associated with historical network issues.

13 . A device, comprising:
one or more processors configured to:
identify a network node that supports a first radio access technology (RAT), wherein the network node is associated with sessions with user equipments (UEs);
calculate a percentage of traffic from the sessions that is associated with the first RAT and not a second RAT;
determine that the network node is associated with a score based on the percentage of traffic associated with the first RAT not satisfying a threshold;
collect, based on the network node being associated with the score, device data associated with the UEs, wherein the device data is associated with signals corresponding to a cell addition procedure;
determine a network issue associated with the network node based on the device data;
determine, via a machine learning (ML) model, a root cause for the network issue; and
transmit an indication of the root cause.

14 . The device of claim 13 , wherein:
the signals corresponding to the cell addition procedure include signals associated with a first radio resource control (RRC) reconfiguration request with a measurement identifier and a band requirement; or the signals corresponding to the cell addition procedure include signals associated with a measurement report for a measurement identifier.

15 . The device of claim 13 , wherein the signals corresponding to the cell addition procedure include signals associated with a second radio resource control (RRC) reconfiguration request with a physical cell identifier (PCI) and a absolute radio-frequency channel number (ARFCN).

16 . The device of claim 13 , wherein the signals corresponding to the cell addition procedure include signals associated with a successful random access channel (RACH) attach message.

17 . The device of claim 13 , wherein the signals corresponding to the cell addition procedure include signals associated with a connection release message.

18 . A non-transitory computer-readable medium storing a set of instructions, the set of instructions comprising:
one or more instructions that, when executed by one or more processors of a server, cause the server to:
identify a network node that supports a first radio access technology (RAT), wherein the network node is associated with sessions with user equipments (UEs);
calculate a percentage of traffic from the sessions that is associated with the first RAT and not a second RAT;
determine that the network node is associated with a score based on the percentage of traffic associated with the first RAT not satisfying a threshold;
collect, based on the network node being associated with the score, device data associated with the UEs, wherein the device data is associated with signals corresponding to a cell addition procedure;
determine a network issue associated with the network node based on the device data;
determine, via a machine learning (ML) model, a root cause for the network issue; and
transmit an indication of the root cause.

19 . The non-transitory computer-readable medium of claim 18 , wherein the network issue is associated with a frequent release of connections associated with the UEs, and the frequent release occurs when the connections are released within an amount of time after a time of connection establishment that satisfies a fourth threshold.

20 . The non-transitory computer-readable medium of claim 18 , wherein the one or more instructions, when executed by the one or more processors, further cause the server to:
determine, via the ML model, the determined root cause for the network issue based on a comparison of features associated with the network issue and features associated with historical network issues.

Assignees

• Verizon Patent & Licensing Inc

Inventors

• Su Le
• Chen Yiming
• Lu Wenyuan
• Wan Yu
• Ruffini Michael P

Classifications

• H04L43/062

  related to network traffic · CPC title

• H04L43/0823 Primary

  Errors, e.g. transmission errors · CPC title

• H04L41/0631 Primary

  using root cause analysis; using analysis of correlation between notifications, alarms or events based on decision criteria, e.g. hierarchy, tree or time analysis · CPC title