US-12621685-B2 - Open radio access network maintenance applications

Abstract

A disclosed method may include (i) building, based on telemetry data from an open radio access network, a machine learning model that predicts when a candidate distributed unit within the open radio access network will experience a failure, (ii) detect, by applying the machine learning model that predicts when the candidate distributed unit will shut down, that a specific distributed unit will experience a specific failure, and (iii) perform, in response to detecting that the specific distributed unit will experience the specific failure, a remedial action that addresses the specific failure. Related systems and computer-readable mediums are further disclosed.

Inventors

• Premchand Chandran
• Gurpreet Sohi
• Julio Armenta

Assignees

• Boost SubscriberCo L.L.C.

Dates

Publication Date

20260505

Application Date

20230622

Claims (20)

1. 1 . A method comprising: building, based on telemetry data from an open radio access network of a cellular telecommunication network, a machine learning model that predicts when a candidate distributed unit within the open radio access network will experience a failure; detecting, by applying the machine learning model that predicts when the candidate distributed unit will experience a failure, that a specific distributed unit will experience a specific failure; and performing, in response to detecting that the specific distributed unit will experience the specific failure, a remedial action that addresses the specific failure failure; wherein a base station of the cellular telecommunication network comprises the specific distributed unit.
2. 2 . The method of claim 1 , wherein the remedial action comprises gracefully shutting down the specific distributed unit by handing over at least one telephone call to another distributed unit.
3. 3 . The method of claim 2 , wherein the machine learning model: specifies a root cause for the specific failure; and specifies the remedial action to be performed in response to the root cause.
4. 4 . The method of claim 2 , wherein gracefully shutting down the specific distributed unit comprises toggling a graceful shutdown flag that triggers a handover procedure for handing over telephone calls to another distributed unit.
5. 5 . The method of claim 4 , wherein the handover procedure prioritizes emergency telephone calls over non-emergency calls.
6. 6 . The method of claim 2 , wherein the remedial action comprises preventing the specific distributed unit from accepting inbound telephone calls during a period of gracefully shutting down.
7. 7 . The method of claim 1 , wherein a cellular telecommunications network comprises the open radio access network.
8. 8 . The method of claim 1 , wherein a 5G telecommunications network comprises the open radio access network.
9. 9 . The method of claim 1 , wherein a gNodeB of the cellular telecommunication network comprises the specific distributed unit.
10. 10 . The method of claim 1 , wherein a radio access network intelligent controller applies the machine learning model to detect that the specific distributed unit will experience the specific failure.
11. 11 . A system comprising: a physical computing processor; and a non-transitory computer-readable medium encoding instructions that, when executed by the physical computing processor, cause a computing device to perform operations comprising: building, based on telemetry data from an open radio access network of a cellular telecommunication network, a machine learning model that predicts when a candidate distributed unit within the open radio access network will experience a failure; detecting, by applying the machine learning model that predicts when the candidate distributed unit will experience a failure, that a specific distributed unit will experience a specific failure; and performing, in response to detecting that the specific distributed unit will experience the specific failure, a remedial action that addresses the specific failure; wherein a base station of the cellular telecommunication network comprises the specific distributed unit.
12. 12 . The system of claim 11 , wherein the remedial action comprises gracefully shutting down the specific distributed unit by handing over at least one telephone call to another distributed unit.
13. 13 . The system of claim 12 , wherein the machine learning model: specifies a root cause for the specific failure; and specifies the remedial action to be performed in response to the root cause.
14. 14 . The system of claim 12 , wherein gracefully shutting down the specific distributed unit comprises toggling a graceful shutdown flag that triggers a handover procedure for handing over telephone calls to another distributed unit.
15. 15 . The system of claim 14 , wherein the handover procedure prioritizes emergency telephone calls over non-emergency calls.
16. 16 . The system of claim 12 , wherein the remedial action comprises preventing the specific distributed unit from accepting inbound telephone calls during a period of gracefully shutting down.
17. 17 . The system of claim 11 , wherein the remedial action comprises: determining that a graceful shutdown of the specific distributed unit is not available; and sending a notification to a network operations center system requesting a prioritized technician dispatch to remediate the specific distributed unit.
18. 18 . The system of claim 11 , wherein the telemetry data comprises at least two of performance management data, fault management data, and log data.
19. 19 . The system of claim 11 , wherein the telemetry data is continuously streamed from the open radio access network to a centralized data platform.
20. 20 . A non-transitory computer-readable medium encoding instructions that, when executed by at least one physical processor of a computing device, cause the computing device to perform operations comprising: building, based on telemetry data from an open radio access network of a cellular telecommunication network, a machine learning model that predicts when a candidate distributed unit within the open radio access network will experience a failure; detecting, by applying the machine learning model that predicts when the candidate distributed unit will experience a failure, that a specific distributed unit will experience a specific failure; and performing, in response to detecting that the specific distributed unit will experience the specific failure, a remedial action that addresses the specific failure; wherein a base station of the cellular telecommunication network comprises the specific distributed unit.

Description

BRIEF SUMMARY This disclosure is generally directed to open radio access network maintenance applications. In one example, a method may include (i) building, based on telemetry data from an open radio access network, a machine learning model that predicts when a candidate distributed unit within the open radio access network will experience a failure, (ii) detecting, by applying the machine learning model that predicts when the candidate distributed unit will shut down, that a specific distributed unit will experience a specific failure, and (ii) performing, in response to detecting that the specific distributed unit will experience the specific failure, a remedial action that addresses the specific failure. In some examples, the remedial action comprises gracefully shutting down the specific distributed unit by handing over at least one telephone call to another distributed unit. In some examples, the machine learning model specifies a root cause for the specific failure and specifies the remedial action to be performed in response to the root cause. In some examples, gracefully shutting down the specific distributed unit comprises toggling a graceful shutdown flag that triggers a handover procedure for handing over telephone calls to another distributed unit. In some examples, the handover procedure prioritizes emergency telephone calls over non-emergency calls. In some examples, the remedial action comprises preventing the specific distributed unit from accepting inbound telephone calls during a period of gracefully shutting down. In some examples, the remedial action comprises determining that a graceful shutdown of the specific distributed unit is not available, and then sending a notification to a network operations center system requesting a prioritized technician dispatch to remediate the specific distributed unit. In some examples, the telemetry data comprises at least two of performance management data, fault management data, and log data. In some examples, the telemetry data is continuously streamed from the open radio access network to a centralized data platform. In some examples, a radio access network intelligent controller applies the machine learning model to detect that the specific distributed unit will experience the specific failure. A corresponding system may include a physical computing processor and a non-transitory computer-readable medium encoding instructions that, when executed by the physical computing processor, cause a computing device to perform operations comprising (i) building, based on telemetry data from an open radio access network, a machine learning model that predicts when a candidate distributed unit within the open radio access network will experience a failure, (ii) detecting, by applying the machine learning model that predicts when the candidate distributed unit will shut down, that a specific distributed unit will experience a specific failure, and (ii) performing, in response to detecting that the specific distributed unit will experience the specific failure, a remedial action that addresses the specific failure. A non-transitory computer-readable medium may encode instructions that, when executed by at least one physical processor of a computing device, cause the computing device to perform operations comprising (i) building, based on telemetry data from an open radio access network, a machine learning model that predicts when a candidate distributed unit within the open radio access network will experience a failure, (ii) detecting, by applying the machine learning model that predicts when the candidate distributed unit will shut down, that a specific distributed unit will experience a specific failure, and (ii) performing, in response to detecting that the specific distributed unit will experience the specific failure, a remedial action that addresses the specific failure. Another example method may include (i) detecting that a software package has become available to be applied within an open radio access network, (ii) detecting, through autonomous monitoring of the open radio access network, that utilization at a set of telecommunications sites is sufficiently low to trigger graceful shutdown and upgrade procedures, (iii) shutting down gracefully the set of telecommunications sites by handing over at least one call to another telecommunications site servicing a common area, (iv) upgrading autonomously the set of telecommunications sites by applying the software package to the open radio access network after shutting down gracefully the set of telecommunications sites. In some examples, the operations may further include restoring the set of telecommunications sites to active functionality in a manner that avoids dropping at least one telephone call. In some examples, the operations may further include executing a sorting algorithm to sort the set of telecommunications sites from a larger set of telecommunications sites due to members of the set of teleco