EP-4740540-A1 - MACHINE LEARNING-BASED CONFLICT MITIGATION FOR XAPPS

Abstract

Architectures and techniques are described that can provide conflict mitigation techniques for xApp that are executed on a near-real-time radio access network intelligent controller (near-RT RIC). A first, deep learning machine learning model can be employed during registration of the xApps, which can identify potential conflicts and flag those potentially conflicting xApps. A second machine learning model can be employed during run time of the xApps, which can identify whether a control message from a given flagged xApp conflicts with a configuration applied to an E2 node by another flagged xApp based on the current network state. Conflicts can be mitigated based on a priority between the two conflicting xApps.

Inventors

• ABOUZEID, MOHAMED
• MANSOUR, Marwan
• SELEEM, Abdelrahman Mohammed

Assignees

• Dell Products, L.P.

Dates

Publication Date

20260513

Application Date

20231028

Claims (20)

1. 1. A device, comprising: a processor; and a memory that stores executable instructions that, when executed by the processor, facilitate performance of operations, comprising: receiving, from a first xApp, of a group of xApps that is registered to a near real time radio access network intelligent controller, a control message that is indicative of a first action to control a first parameter of an E2 node of an open radio access network; in response to a determination that the first xApp has been flagged as a conflict candidate according to a first process using a first machine learning model, receiving active actions data indicative of a second action to control a second parameter of the E2 node that was previously applied to the E2 node by a second xApp of the group of xApps; and based on the active actions data, determining, according to a second process using a second machine learning model, that the first action and the second action conflict.
2. 2. The device of claim 1, wherein the active actions data is received from a data store of the near real time radio access network intelligent controller, the data store configured to store: flag data indicative of conflict candidates, comprising the conflict candidate, determined using the first machine learning model from among the group of xApps registered to the near real time radio access network intelligent controller; history data indicative of actions, comprising the second action, that have been applied to the E2 node; parameter data indicative of parameters, comprising the first parameter and the second parameter, controlled by the group of xApps; category data indicative of a category of a member of the group of xApps; or priority data indicative of a priority of the member of the group of xApps.
3. 3. The device of claim 2, wherein the operations further comprise: receiving the category data and the priority data in response to a registration procedure performed on the member of the group of xApps to register the member to the near real time radio access network intelligent controller; and storing the category data and the priority data to the data store.
4. 4. The device of claim 3, wherein the registration procedure further comprises determining the parameter data and storing the parameter data to the data store.
5. 5. The device of claim 3, wherein the operations further comprise determining the flag data and storing the flag data to the data store.
6. 6. The device of claim 5, wherein the flag data is determined in response to inputting various portions of the parameter data and the category data to the first machine learning model.
7. 7. The device of claim 2, wherein the operations further comprise filtering the history data based on a configurable time window since an action of the actions was applied to the E2 node.
8. 8. The device of claim 2, wherein the operations further comprise filtering the conflict candidates by conflict pairing data indicative of xApps of the group of xApps that were determined to potentially conflict with the first xApp.
9. 9. The device of claim 1, wherein the first action and the second action are determined to conflict based at least in part on a result of inputting the first action and the active actions data to the second machine learning model.
10. 10. The device of claim 1, wherein the operations further comprise, in response to a determination that the first action and the second action conflict, performing a conflict mitigation procedure.
11. 11. The device of claim 10, wherein the conflict mitigation procedure prevents the control message from being delivered to the E2 node in response to a determination that the priority data indicates the second xApp has a higher priority than the first xApp.
12. 12. The device of claim 10, wherein the conflict mitigation procedure instructs the E2 node to be configured according to the control message in response to a determination that the priority data indicates the first xApp has a higher priority than the second xApp.
13. 13. A non-transitory computer- readable medium comprising instructions that, in response to execution, cause a system comprising a processor to perform operations, comprising: receiving, from a first xApp, of a group of xApps that is registered to a near real time radio access network intelligent controller, a control message that is indicative of a first action to control a first parameter of an E2 node of an open radio access network; in response to a determination that the first xApp has been flagged as a conflict candidate according to a first machine learning model, receiving active actions data indicative of a second action to control a second parameter of the E2 node that was previously applied to the E2 node by a second xApp of the group of xApps; based on the active actions data, determining, according to a second machine learning model, that the first action and the second action conflict; and based on a comparison of first priority data assigned to the first xApp and second priority data assigned to the second xApp, determining whether to transmit the control message to the E2 node.
14. 14. The non-transitory computer-readable medium of claim 13, The device of claim 1 , wherein the active actions data is received from a data store of the near real time radio access network intelligent controller, the data store configured to store: flag data indicative of conflict candidates, comprising the conflict candidate, determined according to the first machine learning model from among the group of xApps registered to the near real time radio access network intelligent controller; history data indicative of actions, comprising the second action, that have been applied to the E2 node; parameter data indicative of parameters, comprising the first parameter and the second parameter, controlled by a member of the group of xApps; category data indicative of a category of the member of the group of xApps; or priority data, comprising the first priority data and the second priority data, that is indicative of a priority of the member of the group of xApps.
15. 15. The non-transitory computer-readable medium of claim 14, wherein the operations further comprise: receiving the category data and the priority data in response to a registration procedure performed on the member of the group of xApps to register the member to the near real time radio access network intelligent controller, and storing the category data and the priority data to the data store.
16. 16. The non-transitory computer-readable medium of claim 14, wherein the operations further comprise determining the flag data and storing the flag data to the data store, and wherein the flag data is determined in response to inputting various portions of the parameter data and the category data to the first machine learning model.
17. 17. A method, comprising: receiving, by a device comprising a processor, a control message from a first xApp, of a group of xApps that is registered to a near real time radio access network intelligent controller, wherein the control message is indicative of a first action to control a first configuration of an E2 node of an open radio access network; determining, by the device, that the first xApp has been flagged as a conflict candidate based on a first output from a first machine learning model; receiving, by the device, active actions data indicative of a second action to control a second configuration of the E2 node that was previously applied to the E2 node by a second xApp of the group of xApps; and based on a second output from the active actions data being input to a second machine learning model, determining, by the device, that the first action and the second action conflict.
18. 18. The method of claim 17, further comprising, in response to the determining that the first action and the second action conflict, performing, by the device, a conflict mitigation procedure that compares first priority data of the first xApp to second priority data of the second xApp.
19. 19. The method of claim 18, further comprising, in response to a determination that the first priority data has a higher priority than the second priority data, permitting, by the device, the control message to be transmitted to the E2 node.
20. 20. The method of claim 18, further comprising, in response to a determination that the second priority data has a higher priority than the first priority data, rejecting, by the device, the control message to prevent the first configuration from being applied to the E2 node.

Description

Title: MACHINE LEARNING -BAS ED CONFLICT MITIGATION FOR xAPPS Inventors: Abdelrahman Seleem, Marwan Mansour, and Mohamed Abouzeid RELATED APPLICATION [0001] This application claims priority to U.S. Non-Provisional Patent Application No. 18/347,796, filed July 6, 2023, and entitled “MACHINE LEARNING-BASED CONFLICT MITIGATION FOR xAPPS”, the entirety of which priority application is hereby incorporated by reference herein. BACKGROUND [0002] Open Radio Access Network (O-RAN) is a promising technology that enables network operators to easily integrate different components from different vendors by suggesting new open interfaces and architectures. O-RAN introduces the intelligence of a radio access network (RAN) through a Near-Realtime RAN Intelligent Controller (Near-RT RIC) and Non-Realtime RAN Intelligent Controller (Non-RT RIC) which can enable different vendors to deploy different xApps and rApps to improve network performance in different network slices. For example, an xApp that is deployed on a Near-RT RIC can be purposed to change network configuration elements (e.g., parameters of an E2 Node) based on network metrics data or other data. BRIEF DESCRIPTION OF THE DRAWINGS [0003] Numerous aspects, embodiments, objects, and advantages of the present embodiments will be apparent upon consideration of the following detailed description, taken in conjunction with the accompanying drawings, in which like reference characters refer to like parts throughout, and in which: [0004] FIG. 1 A depicts a schematic block diagram 100 A illustrating an example direct conflict between xApps in accordance with certain embodiments of this disclosure; [0005] FIG. 1 B depicts a schematic block diagram 100B illustrating an example indirect conflict between xApps in accordance with certain embodiments of this disclosure; [0006] FIG. 2 depicts a schematic block diagram illustrating an example system 200 that can utilize machine learning techniques to mitigate control action conflicts that arise between xApps executing on a near-RT RIC in accordance with certain embodiments of this disclosure; [0007] FIG. 3 depicts a schematic block diagram 300 illustrating additional aspects or elements of conflict predictor 214 in accordance with certain embodiments of this disclosure; [0008] FIG. 4 illustrates a schematic block diagram 400 illustrating additional aspects or elements of conflict detector 216 in accordance with certain embodiments of this disclosure; [0009] FIG. 5 depicts a schematic block diagram 500 illustrating additional aspects or elements of conflict mitigator 218 in accordance with certain embodiments of this disclosure; [0010] FIG. 6 depicts a schematic block diagram illustrating an example device 600 that, during registration time, can identify potential conflicts between xApps in accordance with certain embodiments of this disclosure; [0011] FIG. 7 depicts a schematic block diagram illustrating an example device 700 that, during run time, can determine conflicts between xApps from among the candidate conflicts in accordance with certain embodiments of this disclosure; [0012] FIG. 8 depicts a call flow diagram 800 illustrating example conflict mitigation techniques in the context of call flows between xApps and platform functions 802 of a near-RT RIC in accordance with certain embodiments of this disclosure; [0013] FIG. 9 illustrates an example method that can provide conflict mitigation techniques via registration time and run time functions in accordance with certain embodiments of this disclosure; [0014] FIG. 10 illustrates an example method that can provide techniques for conflict mitigation between xApps registered to a near-RT RIC in accordance with certain embodiments of this disclosure; [0015] FIG. 11 illustrates an example method that can provide for additional aspect or elements in connection with providing techniques for conflict mitigation between xApps registered to a near-RT RIC in accordance with certain embodiments of this disclosure; [0016] FIG. 12 illustrates a block diagram of an example distributed file storage system that employs tiered cloud storage in accordance with certain embodiments of this disclosure; and [0017] FIG. 13 illustrates an example block diagram of a computer operable to execute certain embodiments of this disclosure. DETAILED DESCRIPTION OVERVIEW [0018] The disclosed subject matter is now described with reference to the drawings, wherein like reference numerals are used to refer to like elements throughout. In the following description, for purposes of explanation, numerous specific details are set forth in order to provide a thorough understanding of the disclosed subject matter. It may be evident, however, that the disclosed subject matter may be practiced without these specific details. In other instances, well-known structures and devices are shown in block diagram form in order to facilitate describing the disclosed subject matter. [0019] In an 0-RAN deployment, the near-R