Network performance evaluation using AI-based network cloning

Patent metadata

Field	Value
Publication number	US-12598109-B2
Application number	US-202318497319-A
Country	US
Kind code	B2
Filing date	Oct 30, 2023
Priority date	Jan 3, 2023
Publication date	Apr 7, 2026
Grant date	Apr 7, 2026

Abstract

Embodiments herein disclose a method and network apparatus for network performance evaluation using AI-based network cloning. The method includes constructing one or more AI-based network clones of one or more network nodes. The one or more AI-based network clones mimics a data pattern and cell behavior of the one or more network nodes. Further, the method includes receiving a solution predicted by an AI server to mitigate one or more problems associated with one or more services of the one or more network nodes. Further, the method includes evaluating a performance of the one or more AI-based network clones by deploying the solution on the one or more AI-based network clones. Further, the method includes sending the solution to the one or more network nodes for deployment or retraining based on the performance of the one or more AI-based network clones.

First claim

What is claimed is:

1 . A method for network performance evaluation using artificial intelligence-based (AI-based) network cloning by a network apparatus, the method comprising:
constructing at least one AI-based network clone of at least one network node in a wireless network, wherein the at least one AI-based network clone is configured to mimic a data pattern and cell behavior of the at least one network node; receiving a solution predicted by an AI server to mitigate at least one network performance problem associated with at least one service of the at least one network node; evaluating a performance of the at least one AI-based network clone by deploying the solution on the at least one AI-based network clone; determining whether the performance of the at least one AI-based network clone meets a Service Level Agreement (SLA) associated with at least one service of the at least one network node; and sending the solution to the at least one network node for deployment based on the performance of the at least one AI-based network clone meeting the SLA associated with at least one service of the at least one network node; and sending a feedback including the performance of the solution to the AI server for retraining based on the performance of the at least one AI-based network clone not meeting the SLA associated with at least one service of the at least one network node.

2 . The method of claim 1 , wherein constructing the at least one AI-based network clone of the at least one network node in the wireless network comprises:
receiving a request to create the at least one AI-based network clone from the AI server upon detecting the at least one problem associated with at least one service of the at least one network node; obtaining real time data from the at least one network node, wherein the real time data is associated with to the at least one service of the at least one network node to be evaluated; generating the data pattern of the at least one service to be evaluated by applying a second AI model on the real time data, wherein the data pattern comprises at least one of a traffic pattern, a number of users, an amount of data traffic for different services; generating the cell behavior of the at least one service to be evaluated by applying a third AI model the real time data; creating the at least one AI-based network clone of the at least one real network node in the wireless network based on at least one of the data pattern and cell behavior.

3 . The method of claim 1 , wherein the solution to mitigate the at least one problem associated with the at least one service of the at least one real network node is predicted by the AI server by applying a first AI model on real time data obtained from the at least one network node.

4 . The method of claim 1 , further comprising deconstructing the at least one AI-based network clone of the at least one real network node based on the solution being sent to the at least one real network node for deployment.

5 . The method of claim 1 , wherein the network apparatus is deployed in at least one of an independent network server, an open radio access network (ORAN) server, and a self organizing network (SON) server.

6 . A network apparatus for network performance evaluation using artificial intelligence-based (AI-based) network cloning, the network apparatus comprising:
memory storing instructions; and at least one processor comprising processor circuitry coupled to the memory, wherein the instructions, when executed by the at least one processor, individually and/or collectively, cause the network apparatus to:
construct at least one AI-based network clone of at least one network node in a wireless network, wherein the at least one AI-based network clone is configured to mimic a data pattern and cell behavior of the at least one network node;
receive a solution predicted by an AI server to mitigate at least one network performance problem associated with at least one service of the at least one network node;
evaluate a performance of the at least one AI-based network clone by deploying the solution on the at least one AI-based network clone;
determine whether the performance of the at least one AI-based network clone meets a service level agreement (SLA) associated with at least one service of the at least one network node;
send the solution to the at least one network node for deployment based on the performance of the at least one AI-based network clone meeting the SLA associated with at least one service of the at least one network node; and
send a feedback including the performance of the solution to the AI server for retraining based on the performance of the at least one AI-based network clone not meeting the SLA associated with at least one service of the at least one network node.

7 . The network apparatus of claim 6 , wherein for constructing the at least one AI-based network clone of the at least one network node in the wireless network, the instructions, when executed by the at least one processor, individually and/or collectively, cause the network apparatus to:
receive a request to create the at least one AI-based network clone from the AI server upon detecting the at least one problem associated with at least one service of the at least one network node; obtain real time data from the at least one network node, wherein the real time data is associated with to the at least one service of the at least one network node to be evaluated; generate the data pattern of the at least one service to be evaluated by applying a second AI model on the real time data, wherein the data pattern comprises at least one of a traffic pattern, a number of users, an amount of data traffic for different services; generate the cell behavior of the at least one service to be evaluated by applying a third AI model the real time data; create the at least one AI-based network clone of the at least one real network node in the wireless network based on at least one of the data pattern and cell behavior.

8 . The network apparatus of claim 6 , wherein the solution to mitigate the at least one problem associated with the at least one service of the at least one real network node is predicted by the AI server by applying a first AI model on real time data obtained from the at least one network node.

9 . The network apparatus of claim 6 , wherein the instructions, when executed by the at least one processor, individually and/or collectively, cause the electronic device to deconstruct the at least one AI-based network clone of the at least one real network node based on the solution being sent to the at least one real network node for deployment.

10 . The network apparatus of claim 6 , wherein the network apparatus is deployed in at least one of an independent network server, an open radio access network (ORAN) server, and a self organizing network (SON) server.

11 . A non-transitory computer-readable storage medium storing instructions which, when executed by at least one processor comprising processor circuitry of a network apparatus, causes the network apparatus to perform operations, the operations comprising:
constructing at least one artificial intelligence-based (AI-based) network clone of at least one network node in a wireless network, wherein the at least one AI-based network clone is configured to mimic a data pattern and cell behavior of the at least one network node; receiving a solution predicted by an AI server to mitigate at least one network performance problem associated with at least one service of the at least one network node; evaluating a performance of the at least one AI-based network clone by deploying

Assignees

• Samsung Electronics Co Ltd

Inventors

• Singh Sukhdeep
• Thaliath Joseph
• Jaisawal Sandeep Kumar
• Yoon Seungil
• Jain Ashish
• Bhat Avinash
• Thangavel Ganesh Kumar

Classifications

• H04L41/145

  involving simulating, designing, planning or modelling of a network · CPC title

• H04L41/16 Primary

  using machine learning or artificial intelligence · CPC title

• H04L41/5016

  based on statistics of service availability, e.g. in percentage or over a given time · CPC title

• H04L41/5019

  Ensuring fulfilment of SLA · CPC title

• H04L41/40

  using virtualisation of network functions or resources, e.g. SDN or NFV entities · CPC title