CN-121986468-A - Network management operations using artificial intelligence virtual representations

Abstract

A computer-implemented method performed by a node for performing network management operations for a telecommunications network using AI virtual representations of at least a portion of the telecommunications network is provided. The method includes receiving (802) inputs from the ML model, the inputs being based on candidate actions to be taken in the telecommunications network for observation in the telecommunications network. The method further includes initiating (810) an AI virtual representation to obtain a recommended action from the candidate actions based on the inputs, and determining (812) a first change in at least the portion of the telecommunications network resulting from performing the respective candidate action, and (ii) a second change in KPIs of the telecommunications network resulting from performing the respective candidate action. The method further includes providing (814) an output including a recommended action to be taken in the telecommunications network based on the first change and the second change.

Inventors

• FARRELL PADDY
• J. o 'm la.
• S. Salma

Assignees

• 瑞典爱立信有限公司

Dates

Publication Date

20260505

Application Date

20230928

Claims (20)

1. 1. A computer-implemented method performed by a node for performing network management operations for a telecommunications network using an artificial intelligence, AI, virtual representation of at least a portion of the telecommunications network, the method comprising: Receiving (802) a plurality of inputs from a plurality of machine learning, ML, models, the plurality of inputs being based on a plurality of candidate actions to be taken in the telecommunications network for observation in the telecommunications network; Based on the plurality of inputs, initiating (810) the AI virtual representation to obtain a recommended action from the plurality of candidate actions; Determining (812) using the AI virtual representation (i) a first change in at least the portion of the telecommunications network resulting from performing a respective candidate action on at least the portion of the telecommunications network and (ii) a second change in a key performance indicator, KPI, of the telecommunications network resulting from performing the respective candidate action on the telecommunications network, and Providing (814) an output from the AI virtual representation including the recommended action from the plurality of candidate actions to be taken in the telecommunications network based on the first change and the second change.
2. 2. The method of claim 1, wherein the plurality of inputs from the plurality of ML models comprises one or more of: (i) A first input from a first machine learning, ML, model comprising a relationship between data related to the observations from at least one of the telecommunications network and a source external to the telecommunications network, (Ii) A second input from a second ML model including an estimate of the first and second variations of the respective candidate actions, and (Iii) A third input comprising feedback related to the accuracy of the output of the AI virtual representation.
3. 3. The method of any of claims 1-2, wherein the plurality of inputs is based on at least one of (i) a first input initiated from a user interface regarding the plurality of candidate actions, and (ii) a second input regarding the plurality of candidate actions designated for or discovered by the AI virtual representation.
4. 4. A method according to any one of claims 1 to 3, further comprising: Synchronizing (800) at least the portion of the telecommunications network and the AI virtual representation in real time or near real time.
5. 5. The method of any of claims 2 to 4, wherein the first ML model comprises a graph-based ML model that learns dependencies in data related to the observations between functions, states, processes and/or assets in at least the portion of the telecommunications network.
6. 6. The method of any of claims 2 to 5, wherein the second ML model comprises a predictive model that provides the estimate in near real-time.
7. 7. The method of any of claims 1 to 6, wherein the output is from a fourth ML model comprising a reinforcement learning model.
8. 8. The method of any of claims 2 to 7, wherein the relationship from the first ML model is identified based on (i) identifying data from the observations that is relevant to the plurality of candidate actions, (ii) determining that the data is available, (iii) determining dependencies between the data, and (iv) building a plurality of states and profiles for the dependencies.
9. 9. The method of any of claims 1 to 8, further comprising: determining (804) whether data related to the plurality of inputs is available for a corresponding ML model of the plurality of ML models, and One of (i) deploying the plurality of ML models when the data is available, and (ii) initiating obtaining the data when the data is not available is performed (806).
10. 10. The method of any of claims 1 to9, further comprising: at least one ML model of the plurality of ML models is trained (808).
11. 11. The method of claim 10, wherein the training (808) comprises (i) determining whether the at least one of the plurality of ML models is the most current version, and (ii) determining whether training data is available when the at least one of the plurality of ML models is the most current version, or (iii) obtaining the most current version when the at least one of the plurality of ML models is not the most current version.
12. 12. The method of claim 11, wherein the training (808) further comprises (i) validating an infrastructure for the training, (ii) setting at least one hyper-parameter for the at least one ML model of the plurality of ML models, and (iii) initiating the training.
13. 13. The method of any of claims 1-12, wherein providing (814) the output is based on setting the respective candidate action according to the respective first and second variations and back-propagating the respective candidate action to derive the recommended action.
14. 14. The method of any one of claims 9 to 13, wherein obtaining the data comprises using a third ML model to retrieve the data via a data proxy.
15. 15. The method of claim 14, wherein obtaining the data comprises (i) identifying a data source of the data, and (ii) establishing a data proxy to obtain the data.
16. 16. The method of any one of claims 1 to 15, further comprising: a display of at least one of the first change and the second change is initiated (816).
17. 17. The method of any one of claims 1 to 16, further comprising: the AI virtual representation and the output are saved (818).
18. 18. The method of any of claims 1-17, wherein the AI virtual representation is located in a cloud-based node.
19. 19. The method of any of claims 1-18, wherein the AI virtual representations are distributed over a plurality of cloud-based nodes.
20. 20. The method of any of claims 1-19, wherein the node comprises a component comprising rApp in a service and administration SMO platform.

Description

Network management operations using artificial intelligence virtual representations Technical Field The present disclosure relates to wireless communication systems, and more particularly, to a computer-implemented method of performing network management operations for a telecommunications network by a node using an Artificial Intelligence (AI) virtual representation of at least a portion of the telecommunications network. Background A network virtual representation (e.g., network digital twinning) refers to a virtual or digital representation/copy of an actual network asset or process that is periodically synchronized with the network and its environment. The virtual representation may be a key element in network design, operation, and maintenance, with the ability to, for example, provide an accurate up-to-date virtual representation of the network and its surroundings consistent with predicted impact on network changes due to internal or external influencing factors. Disclosure of Invention For example, some methods may use virtual representations for manufactured products and for operating health and lifetime of manufactured products, or for manufacturing operations involving virtual factories or workshops. Such methods do not include virtual representations of a network (e.g., a telecommunications network) or a portion of a network. Additionally, they also do not relate to management operations in the network, in particular management operations based on multiple inputs, and may not relate to the construction, synchronization and/or self-learning of virtual representations of the network, in particular from a network management point of view. The operational management of a network, such as a Radio Access Network (RAN), can be complex. Some methods for operation management of a network may include attempting to improve performance or optimize features of use in some manner, such as radio sleep or hibernation features that reduce energy usage. Such features may have a single purpose (e.g., reduce energy costs) and may not take into account the impact on network functionality and quality, as this is typically controlled by restrictions aimed at constraining the detrimental impact such features have on network quality. Furthermore, it may be difficult for network engineers to activate such features and adjust their limitations to predict and visualize the impact. Furthermore, network functions and management systems may lack the ability to implement zero-touch control for the network, as the actual impact of the change may not be predictable. Certain aspects of the present disclosure and embodiments thereof are capable of providing solutions to these and other challenges. In some embodiments, a computer-implemented method performed by a node for performing network management operations for a telecommunications network using an artificial intelligence, AI, virtual representation of at least a portion of the telecommunications network is provided. The method includes receiving a plurality of inputs from a plurality of machine learning ML models. The plurality of inputs is based on a plurality of candidate actions to be taken in the telecommunications network for observation in the telecommunications network. The method also includes initiating the AI virtual representation to obtain a recommended action from the plurality of candidate actions based on the plurality of inputs. The method further includes determining, using the AI virtual representation, (i) a first change in at least the portion of the telecommunications network resulting from performing a respective candidate action on the at least the portion of the telecommunications network, and (ii) a second change in a key performance indicator, KPI, of the telecommunications network resulting from performing the respective candidate action on the telecommunications network. The method further includes providing, based on the first and second variations, an output from the AI virtual representation including the recommended action from the plurality of candidate actions to be taken in the telecommunications network. According to some embodiments, a node is provided that is configured to perform network management operations for a telecommunication network using an AI virtual representation of at least a portion of the telecommunication network. The node includes processing circuitry and memory coupled with the processing circuitry. The memory includes instructions that, when executed by the processing circuitry, cause the node to perform operations. The operations include receiving a plurality of inputs from a plurality of ML models. The plurality of inputs is based on a plurality of candidate actions to be taken in the telecommunications network for observation in the telecommunications network. The operations also include initiating the AI virtual representation to obtain a recommended action from the plurality of candidate actions based on the plura