IP Library › Granted Patent US 12694330

Granted Patent B2

US 12694330 · App. 17/921,642 · Granted Jul 28, 2026

Method and apparatus for transferring machine learning model parameter

Inventors: Shiqiang Suo (Beijing, CN); Yingmin Wang (Beijing, CN)

Assignee: DATANG MOBILE COMMUNICATIONS EQUIPMENT CO., LTD.

G06N20/00 H04L41/16 H04W80/12

View Patent ↗

Loading inventors, assignments & file history…

Monitor This Case

Get email alerts when status or documents change.

Order Certified Copies

Most orders are placed with the USPTO same day — all within 24 business hours.

Order via The Patent Place →

Pre-filled with this patent's details

Quick Facts

Patent No.: US 12694330
App. No.: 17/921,642
Granted: Jul 28, 2026
Kind: B2

External Links

USPTO Patent Center↗ Full Patent PDF↗

Abstract

A method and apparatus for transferring machine learning model parameters are provided. The method includes: an AI-native service slave management component in a first device receives a machine learning model file and information of a target functional component sent by an AI-native service master management component in a second device; and the AI-native service slave management component distributes the machine learning model file into the target functional component in the first device according to the information of the target functional component.

Claims (46)

1 . A method for transferring machine learning model parameters in a mobile communication system, applied to a first device, comprising:

receiving, by an artificial intelligence (AI)-native service slave management component in the first device, a machine learning model file and information of a target functional component sent by an AI-native service master management component in a second device, wherein the target functional component comprises at least one of a channel estimation component, a signal detection component, a channel decoding component, a channel quality indicator (CQI) measurement component, a synchronization detection component, or a transmitting antenna selection component; and

distributing, by the AI-native service slave management component, the machine learning model file into the target functional component in the first device according to the information of the target functional component;

wherein the method further comprises:

notifying, by the target functional component, the AI-native service slave management component that the machine learning model file has been installed;

applying, by the AI-native service slave management component, for a transmission resource between the first device and the second device from a transmission resource management component in the second device;

applying, by the AI-native service slave management component, for service registration from the AI-native service master management component by using the transmission resource obtained through application; and

launching, by the AI-native service slave management component, a service included in a latest machine learning model installed in the target functional component after obtaining a service registration permission from the AI-native service master management component;

wherein the first device is a terminal device in the mobile communication system, the second device is a base station device in the mobile communication system, and the transmission resource between the first device and the second device is a wireless transmission resource between the terminal device and the base station device; or

wherein the first device is a base station distributed unit in the mobile communication system, the second device is a base station centralized unit in the mobile communication system, and the transmission resource between the first device and the second device is a wired or wireless transmission resource between the base station distributed unit and the base station centralized unit.

2 . The method according to claim 1 , wherein the AI-native service slave management component is located in an application layer of the first device; and the AI-native service master management component is located in an application layer of the second device.

3 . The method according to claim 1 , further comprising:

installing, by the target functional component, the machine learning model file.

4 . A method for transferring machine learning model parameters in a mobile communication system, applied to a second device, comprising:

determining, by an AI-native service master management component in the second device, a target functional component in which a machine learning model is to be deployed in a first device and a corresponding machine learning model file, wherein the target functional component comprises at least one of a channel estimation component, a signal detection component, a channel decoding component, a channel quality indicator (CQI) measurement component, a synchronization detection component, or a transmitting antenna selection component;

applying, by the AI-native service master management component in the second device, for a transmission resource between the first device and the second device from a transmission resource management component in the second device; and

sending, by the AI-native service master management component in the second device, information of the target functional component and the machine learning model file to an AI-native service slave management component in the first device by using the transmission resource obtained by application;

wherein the method further comprises:

receiving, by the AI-native service master management component in the second device, a transmission resource application sent by the AI-native service slave management component in the first device, and allocating the transmission resource between the first device and the second device to the AI-native service slave management component;

receiving, by the AI-native service master management component in the second device, a service registration application sent by the AI-native service slave management component in the first device, and sending a service registration permission to the AI-native service slave management component;

5 . The method according to claim 4 , wherein the AI-native service slave management component is located in an application layer of the first device; and the AI-native service master management component is located in an application layer of the second device.

6 . An apparatus for transferring machine learning model parameters in a mobile communication system, applied to a first device, comprising:

a memory configured to store program instructions;

a processor configured to invoke the program instructions stored in the memory and execute according to an obtained program:

controlling an AI-native service slave management component in the first device to receive a machine learning model file and information of a target functional component sent by an AI-native service master management component in a second device, wherein the target functional component comprises at least one of a channel estimation component, a signal detection component, a channel decoding component, a channel quality indicator (CQI) measurement component, a synchronization detection component, or a transmitting antenna selection component; and

controlling the AI-native service slave management component to distribute the machine learning model file into the target functional component in the first device according to the information of the target functional component,

wherein the processor is further configured to invoke the program instructions stored in the memory and execute according to the obtained program:

controlling the target functional component to notify the AI-native service slave management component that the machine learning model file has been installed;

controlling the AI-native service slave management component to apply for a transmission resource between the first device and the second device from a transmission resource management component in the second device;

controlling the AI-native service slave management component to apply for service registration to the AI-native service master management component by using the transmission resource obtained through application; and

controlling the AI-native service slave management component to launch a service included in a latest machine learning model installed in the target functional component after obtaining a service registration permission from the AI-native service master management component;

7 . The apparatus according to claim 6 , wherein the AI-native service slave management component is located in an application layer of the first device; and the AI-native service master management component is located in an application layer of the second device.

8 . The apparatus according to claim 6 , wherein the processor is further configured to invoke the program instructions stored in the memory and execute according to the obtained program:

controlling the target functional component to install the machine learning model file.

9 . An apparatus for transferring machine learning model parameters in a mobile communication system, applied to a second device, comprising:

a memory configured to store program instructions;

a processor configured to invoke the program instructions stored in the memory and execute according to an obtained program the method of claim 4 .

10 . The apparatus according to claim 9 , wherein the AI-native service slave management component is located in an application layer of the first device; and the AI-native service master management component is located in an application layer of the second device.

11 . The method according to claim 1 , wherein the machine learning model file comprises an executable program file.

12 . The method according to claim 4 , wherein the machine learning model file comprises an executable program file.

13 . The apparatus according to claim 6 , wherein the machine learning model file comprises an executable program file.

14 . The apparatus according to claim 9 , wherein the machine learning model file comprises an executable program file.