CN-121986512-A - Registration and discovery of model training for artificial intelligence at user equipment

Abstract

Certain aspects of the present disclosure provide techniques for registering and discovering User Equipment (UE) model training functions in a core network. An example method for wireless communication by an apparatus includes transmitting, to a first network entity, an indication of a registration Model Training Logic Function (MTLF), the Model Training Logic Function (MTLF) having the ability to train a Machine Learning (ML) model for deployment at one or more User Equipments (UEs), obtaining an indication of training the ML model based at least in part on the indication of registration MTLF, obtaining training data associated with the one or more UEs, training the ML model based at least in part on the training data, and transmitting, to a second network entity, an indication of transmitting the trained ML model to at least one UE.

Inventors

• ZHANG JUAN
• H. Cisimopoulos
• G.B. Horn
• A. Gormi

Assignees

• 高通股份有限公司

Dates

Publication Date

20260505

Application Date

20231010

Claims (18)

1. 1. An apparatus configured for wireless communication, the apparatus comprising: One or more memories, and One or more processors coupled to the one or more memories, the one or more processors configured to cause the apparatus to: Transmitting an indication to a first network entity of a registration Model Training Logic Function (MTLF), the Model Training Logic Function (MTLF) having the capability to train a Machine Learning (ML) model for deployment at one or more User Equipments (UEs); Obtaining an indication to train the ML model based at least in part on the indication to register the MTLF; obtaining training data associated with one or more UEs; Training the ML model based at least in part on the training data, and An indication of transmission of the trained ML model to the at least one UE is transmitted to the second network entity.
2. 2. The apparatus of claim 1, wherein to train the ML model, the one or more processors are configured to cause the apparatus to train the ML model to predict one or more attributes associated with one or more of channel state feedback, beamforming, or positioning.
3. 3. The apparatus of claim 1, wherein to communicate the indication to register the MTLF, the one or more processors are configured to cause the apparatus to communicate a registration request to the first network entity that includes the indication to register the MTLF.
4. 4. The apparatus of claim 3, wherein the registration request further comprises one or more of: MTLF identifier identifying the MTLF; a function identifier identifying one or more functions for which the ML model is used; An analysis identifier identifying the MTLF one or more ML models capable of training; an indication of at least one ML model for which the MTLF is capable of training; an application identifier identifying the MTLF or a network data analysis function (NWDAF) hosting the MTLF, or A vendor identifier identifying the MTLF or the service provider of NWDAF.
5. 5. The apparatus of claim 1, wherein to obtain the indication to train the ML model, the one or more processors are configured to cause the apparatus to obtain the indication to train the ML model from the first network entity.
6. 6. The apparatus of claim 1, wherein the apparatus comprises a third network entity configured to perform a network data analysis function (NWDAF).
7. 7. The apparatus of claim 1, wherein the first network entity comprises a network controller.
8. 8. The apparatus of claim 1, wherein the first network entity is configured to perform a Network Repository Function (NRF).
9. 9. An apparatus configured for wireless communication, the apparatus comprising: One or more memories, and One or more processors coupled to the one or more memories, the one or more processors configured to cause the apparatus to: Transmitting a request to a first network entity to discover one or more Network Function (NF) profiles; Obtaining an indication of the one or more NF profiles from the first network entity, the one or more NF profiles including NF profiles associated with a Model Training Logic Function (MTLF), the NF profiles indicating the MTLF ability to train a Machine Learning (ML) model for deployment at one or more User Equipment (UE), and An indication to train the ML model using the MTLF is communicated based at least in part on the NF profile.
10. 10. The apparatus of claim 9, wherein the indication to train the ML model comprises one or more of an indication of training data to collect, an indication of one or more UEs from which some or all of the training data is to be collected, or an indication of the ML model to train.
11. 11. The apparatus of claim 9, wherein the indication to train the ML model comprises one or more of: a function identifier identifying one or more functions for which the ML model is used; An analysis identifier identifying the MTLF one or more ML models capable of training; Collecting the duration of the training data; a sampling frequency for sampling the training data at one or more UEs; reporting frequency for reporting the training data; a location from which the training data is to be collected; the total number of the one or more UEs from which the training data is to be collected, or One or more characteristics associated with the ML model.
12. 12. The apparatus of claim 9, wherein the request comprises a discovery request for at least one NF profile associated with at least one MTLF.
13. 13. The apparatus of claim 9, wherein the NF profile comprises one or more of: MTLF identifier identifying the MTLF; a function identifier identifying one or more functions for which the ML model is used; An analysis identifier identifying the MTLF one or more ML models capable of training; an indication of at least one ML model for which the MTLF is capable of training; an application identifier identifying the MTLF or a network data analysis function (NWDAF) hosting the MTLF, or A vendor identifier identifying the MTLF or the service provider of NWDAF.
14. 14. The apparatus of claim 9, wherein to communicate the indication to train the ML model, the one or more processors are configured to cause the apparatus to communicate the indication to train the ML model to a second network entity.
15. 15. The apparatus of claim 14, wherein the second network entity is configured to perform a network data analysis function (NWDAF).
16. 16. The apparatus of claim 9, wherein: the device comprises a network controller, and The first network entity is configured to perform a Network Repository Function (NRF).
17. 17. A method of wireless communication by a device, the method comprising: Transmitting an indication to a first network entity of a registration Model Training Logic Function (MTLF), the Model Training Logic Function (MTLF) having the capability to train a Machine Learning (ML) model for deployment at one or more User Equipments (UEs); Obtaining an indication to train the ML model based at least in part on the indication to register the MTLF; obtaining training data associated with one or more UEs; Training the ML model based at least in part on the training data, and An indication of transmission of the trained ML model to the at least one UE is transmitted to the second network entity.
18. 18. A method of wireless communication by a device, the method comprising: Transmitting a request to a first network entity to discover one or more Network Function (NF) profiles; Obtaining an indication of the one or more NF profiles from the first network entity, the one or more NF profiles including NF profiles associated with a Model Training Logic Function (MTLF), the NF profiles indicating the MTLF ability to train a Machine Learning (ML) model for deployment at one or more User Equipment (UE), and An indication to train the ML model using the MTLF is communicated based at least in part on the NF profile.

Description

Registration and discovery of model training for artificial intelligence at user equipment Introduction to the invention Technical Field Aspects of the present disclosure relate to wireless communications, and more particularly, to techniques for managing artificial intelligence for use at user equipment. Description of related Art Wireless communication systems are widely deployed to provide various telecommunication services such as telephony, video, data, messaging, broadcast, or other similar types of services. These wireless communication systems may employ multiple-access techniques that are capable of supporting communication with several users by sharing the available wireless communication system resources with those users. Despite the tremendous technological advances made over the years in wireless communication systems, challenges remain. For example, complex and dynamic environments may still attenuate or block signals between a wireless transmitter and a wireless receiver. Accordingly, there is a continuing desire to improve the technical performance of wireless communication systems, including, for example, improving the speed and data carrying capacity of communications, improving the efficiency of use of shared communication media, reducing the power used by transmitters and receivers in performing communications, improving the reliability of wireless communications, avoiding redundant transmissions and/or receptions and associated processing, improving the coverage area of wireless communications, increasing the number and types of devices that can access the wireless communication system, increasing the ability of different types of devices to communicate with each other, increasing the number and types of wireless communication media available for use, and the like. Accordingly, there is a need for further improvements in wireless communication systems to overcome the foregoing technical challenges and others. Disclosure of Invention One aspect provides a method for wireless communication by an apparatus. The method includes transmitting an indication of a registration Model Training Logic Function (MTLF) to a first network entity, the Model Training Logic Function (MTLF) having the capability to train a Machine Learning (ML) model for deployment at one or more User Equipment (UEs), obtaining an indication of training the ML model based at least in part on the indication of registration MTLF, obtaining training data associated with the one or more UEs, training the ML model based at least in part on the training data, and transmitting an indication of transmitting the trained ML model to at least one UE to a second network entity. Another aspect provides a method for wireless communication by an apparatus. The method includes transmitting a request to a first network entity to discover one or more NF profiles, obtaining an indication of one or more Network Functions (NF) profiles from the first network entity, the one or more Network Functions (NF) profiles including NF profiles associated with MTLF, the NF profile indication MTLF to train capabilities of an ML model for deployment at one or more UEs, and transmitting an indication to train the ML model using MTLF based at least in part on the NF profiles. Another aspect provides a method for wireless communication by an apparatus. The method includes obtaining an indication of registration MTLF from a first network entity, the MTLF having the ability to train an ML model for deployment at one or more UEs, obtaining a request from a second network entity to discover one or more NF profiles, and transmitting an indication of the one or more NF profiles to the second network entity, the one or more NF profiles including an NF profile associated with MTLF, the NF profile indicating MTLF the ability to train the ML model. Other aspects provide one or more devices operable to, configured or otherwise adapted to perform any portion of any of the methods described herein (e.g., such that execution may be performed by only one device or in a distributed manner across multiple devices), one or more non-transitory computer-readable media comprising instructions that, when executed by one or more processors of one or more devices, cause the one or more devices to perform any portion of any of the methods described herein (e.g., such that the instructions may be included in only one computer-readable medium or in a distributed manner across multiple computer-readable media, such that the instructions may be executed by only one processor or in a distributed manner across multiple processors, such that each of the one or more devices may include one or more processors, and/or such that execution may be performed by only one device or more devices, such as in a distributed manner across one or more devices), one or more computer-readable media storing any portion of any of the methods described herein (e.g., such that the one or more computer-readable