Method and apparatus for UE data transfer for ai / ML model training

Abstract

This invention focuses on Standardized transfer of data over UP for UE data collection as part of UE-side model Training. The method utilizes two different approaches one based on UE- initiated PDU Session establishment and another based on Network-initiated UP Tunnel.

Description

Method and apparatus for UE data transfer for AI / ML Model TrainingField

[0001] This invention focuses on how to support the standardized transfer of standardized data over UP for UE data collection to meet requirements for AI / ML for NR air interface operation with UE-side model training.BACKGROUND

[0002] The invention addresses issue related on UP Data collection for AI / ML as documented in 3GPP RAN LSs RP -243316 and RP-242389. The requirements are elaborated in LSs RP-242389 and R2-2411152. All documents are publicly available via 3GPP portal www.3GPP.org.SUMMARY

[0003] The following presents a simplified summary of one or more aspects in order to provide a basic understanding of such aspects. This summary is not an extensive overview of all contemplated aspects and is intended to neither identify key or critical elements of all aspects nor delineate the scope of any or all aspects. Its sole purpose is to present some concepts of one or more aspects in a simplified form as a prelude to the more detailed description that is presented later.

[0004] In an aspect of the disclosure, a method, a computer-readable medium, and an apparatus are provided. A method of Standardized transfer of data over UP for UE data collection as part of UE-side model Training. The method utilizes two different approaches one based on UE-initiated PDU Session establishment and another based on Network-initiated UP Tunnel.

[0005] To the accomplishment of the foregoing and related ends, the one or more aspects comprise the features hereinafter fully described and particularly pointed out in the claims. The following description and the annexed drawings set forth in detail certain illustrative features of the one or more aspects. These features are indicative, however, of but a few of the various ways in which the principles of various aspects may be employed, and this description is intended to include all such aspects and their equivalents.BRIEF DESCRIPTION OF THE DRAWINGSFigure 1. Procedure to transfer collected data from UE (UE-initiated PDU Session)Figure 2. Procedure to transfer collected data from UE (Network-initiated UP Tunnel)DETAILED DESCRIPTION

[0006] The detailed description set forth below in connection with the appended drawings is intended as a description of various configurations and is not intended to represent the only configurations in which the concepts described herein may be practiced. The detailed description includes specific details for the purpose of providing a thorough understanding of various concepts. However, it will be apparent to those skilled in the art that these concepts may be practiced without these specific details. In some instances, well known structures and components are shown in block diagram form in order to avoid obscuring such concepts.

[0007] Figure 1. Procedure to transfer collected data from UE (UE-initiated PDU Session)

[0008] Steps to be taken as part Figure 1 is as follows:

[0009] In Step 0, It is assumed Data Collection and related Configuration Procedure on the UE already initiated by RAN. The data is being logged in the UE internally.

[0010] In Step 1, The Data Transfer Request is triggered from Server for UE-side data collection towards an NF in CN in charge of Data Transfer. Here, we refer to this NF as DTNF. The request may include standardized data types and other filter information that can beprocessed by UE. The request may also include authentication information that enable the DTNF to verify the authenticity of data collected.

[0011] DTNF can be a new NF within CN or a implemented as subset of an existing NF in CN.

[0012] A single instance or multiple instances of DTNF may be deployed in a PLMN. If multiple DTNF instances are deployed, the architecture supports deploying the DTNF as a central NF, as a collection of distributed NF s, or as a combination of both.

[0013] In Step 2a. UE is configured / provisioned with Data Transfer Configuration Information (DTCI). The DTCI includes at least the address (e.g. FQDN) of the data repository within CN where the UE logged data to be transferred to.

[0014] UE is provisioned with Data DTCI from CN. This can be based on static, dynamic polices, N1 NAS signaling or any other forms of in-band or out-of-band signaling.

[0015] DTCI configuration may also include information related to Data Network, DNN,Slicing or other relevant information to enable UE PDU Session establishment to data repository.

[0016] In Step 3a. UE establishes a UP PDU session to the data repository within CN based on configuration information provided in Step 2a. As part of PDU session establishment procedure, DTNF triggers an N2 PDU session request to RAN. RAN and UE perform AN-specific resource setup. RAN responds to DTNF with N2 PDU session response. Data Transfer starts after this step as long as UE has the data collected.

[0017] In Step 4a. Optionally, Data Transfer may start when Data Transfer request (including any standardized data types, other filter information or other authentication information) is conveyed from DTNF to UE.

[0018] In Step 5a. UE initiates the data Transfer to the data repository within CN over the UP PDU session.

[0019] In Step 6. DTNF responds to Server for UE-side data collection on the date file to be ready within the data repository. The Server for UE-side data collection fetches data from repository by using file transfer protocols.

[0020] Figure 2. Procedure to transfer collected data from UE (Network-initiated UP Tunnel)

[0021] Steps to be taken as part Figure 2 is as follows:

[0022] In Step 0, it is assumed Data Collection and related Configuration Procedure on the UE already initiated by RAN. The data is being logged in the UE internally.

[0023] In Step 1, the Data Transfer Request is triggered from Server for UE-side data collection towards DTNF. The request may include standardized data types and other filter information that can be processed by RAN node. The request may also includeauthentication information that enable the DTNF to verify the authenticity of data collected.

[0024] DTNF can be a new NF within CN or a subset of an existing NF in CN.

[0025] A single instance or multiple instances of DTNF may be deployed in a PLMN. If multiple DTNF instances are deployed, the architecture supports deploying the DTNF as a central NF, as a collection of distributed NF s, or as a combination of both.

[0026] In Step 2b, RAN is configured / provisioned with DTCI. The DTCI includes at least the address (e.g., FQDN) of the data repository within CN where the UE log report to be transferred to.

[0027] RAN is provisioned with DTCI from CN. This can be based on static, dynamic polices, configurations, N2 signaling or any other forms of in-band or out-of-band signaling.

[0028] DTCI configuration may also include information related to Data Network, DNN,Slicing or other relevant information to enable RAN / Network to establish UP tunnel to data repository.

[0029] Data repository within CN for transferred data can be in DTNF or any other NF.

[0030] In Step 3b, RAN establishes a UP Tunnel to the data repository within CN based on configuration information provided in Step 2b. RAN also interacts with UE to perform AN-specific resource set up. Data Transfer starts after this step as long as UE has the data collected.

[0031] In Step 4b, Optionally, Data Transfer may start when Data Transfer request (including any standardized data types, other filter information or other authentication information) is conveyed from DTNF to RAN node.

[0032] In Step 5b, UE initiates Data Transfer to the data repository within CN over the UP tunnel.

[0033] In Step 6, DTNF responds to Server for UE-side data collection on the date file to be ready within the data repository. The Server for UE-side data collection fetches data from repository by using file transfer protocols.

Claims

CLAIMS1. Method and apparatus for UE data transfer for AI / ML Model Training compromising of UE-initiated PDU Session establishment and Network-initiated UP Tunnel.

Images