IP Library Granted Patent US 12,499,384
Granted Patent B2
US 12,499,384 · App. 17/795,400 · Granted Dec 16, 2025

User equipment artificial intelligence-machine-learning capability categorization system, method, device, and program

Inventors: Awn Muhammad (Tokyo, JP); Koichiro Kitagawa (Tokyo, JP); Krishnakumar Kesavan (San Mateo, CA); Taewoo Lee (Tokyo, JP)
Assignee: RAKUTEN MOBILE, INC.
G06N20/00G06N5/04
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 12,499,384
App. No.
17/795,400
Granted
Dec 16, 2025
Kind
B2
Abstract

A method, system, apparatus, and non-transitory computer-readable medium for classifying machine-learning capabilities of a user device in a telecommunication network may be provided. The method may be performed by one or more processors, and may include receiving user device capability information from the user device; based on the user device capability information, determining a classification of machine learning capabilities of the user device; and transmitting, to the user device, data associated with a machine learning model based on the classification of the machine learning capabilities of the user device.

Claims (55)

1 . A method for classifying machine-learning capabilities of a user device in a telecommunication network, the method being executed by a processor, and the method comprising:

receiving user device capability information from the user device;

based on the user device capability information, determining a classification of machine-learning capabilities of the user device; and

transmitting, to the user device, data associated with a machine-learning model based on the classification of the machine-learning capabilities of the user device.

2 . The method of claim 1 , wherein the user device capability information includes a plurality of parameters associated with the user device, and wherein each parameter is expressed as a categorical variable.

3 . The method of claim 2 , wherein the plurality of parameters associated with the user device include at least one of a processor type of the user device, a size of available memory, a battery power of the user device, a battery health of the user device, a device type of the user device, or radio frequency hardware capability of the user device.

4 . The method of claim 3 , wherein determining the classification comprises:

evaluating one or more parameters of the user device from the plurality of parameters;

determining a categorical value for each of the one or more parameters; and

based on comparing the categorical value for each of the one or more parameters to a pre-determined set of minimum categorical values, assigning a classification number to the user device based on the categorical value for each of the one or more parameters.

5 . The method of claim 2 , wherein transmitting data associated with a machine-learning model to the user device comprises:

based on the classification of the machine-learning capabilities of the user device, transmitting, to the user device, one of:

training data and model parameters for a full scale training at the user device;

a lightly trained model, a subset of the training data, and the model parameters for a lightweight training at the user device;

a general trained model, a subset of training data, and the model parameters for a specific use case-based update of the general trained model at the user device; and

a trained model for an inference at the user device.

6 . The method of claim 5 , wherein the specific use case-based update for the general trained model is associated with a use case from among a channel-state information (CSI) feedback enhancement, beam management, positioning accuracy, received signal (RS) overhead reduction, load balancing, mobility optimization, or network energy saving.

7 . The method of claim 5 , wherein the lightweight training at the user device includes updating the lightly trained model for a limited number of epochs to achieve an acceptable level of accuracy.

8 . The method of claim 5 , wherein the full scale training at the user device includes generating the machine-learning model using the training data and the model parameters, wherein the generating includes training a model for a large number of epochs to achieve a high level of accuracy.

9 . The method of claim 1 , wherein the classification of the machine-learning capabilities of the user device is indicative of an artificial intelligence model training capacity of the user device.

10 . The method of claim 1 , wherein the classification of the machine-learning capabilities of the user device is indicative of an artificial intelligence model inference capacity of the user device.

11 . The method of claim 1 , wherein the receiving of the user device capability information is in response to receiving a request from the user device.

12 . The method of claim 1 , wherein the receiving of the user device capability information is in response to receiving a request from a network element of the telecommunication network.

13 . An apparatus comprising:

a memory configured to store instructions; and

one or more processors configured to execute the instructions to:

receive user device capability information from the user device;

based on the user device capability information, determine a classification of machine-learning capabilities of the user device; and

transmit, to the user device, data associated with a machine-learning model based on the classification of the machine-learning capabilities of the user device.

14 . The apparatus of claim 13 , wherein the user device capability information includes a plurality of parameters associated with the user device, and wherein each parameter is expressed as a categorical variable.

15 . The apparatus of claim 14 , wherein the determining the classification comprises:

evaluating one or more parameters of the user device from the plurality of parameters;

determining a categorical value for each of the one or more parameters; and

based on comparing the categorical value for each of the one or more parameters to a pre-determined set of minimum categorical values, assigning a classification number to the user device based on the categorical value for each of the one or more parameters.

16 . The apparatus of claim 13 , wherein transmitting data associated with a machine-learning model to the user device comprises:

based on the classification of the machine-learning capabilities of the user device, transmitting, to the user device, one of:

training data and model parameters for a full scale training at the user device;

a lightly trained model, a subset of the training data, and the model parameters for a lightweight training at the user device;

a general trained model, a subset of training data, and the model parameters for a specific use case-based update of the general trained model at the user device; and

a trained model for an inference at the user device.

17 . The apparatus of claim 16 , wherein the lightweight training at the user device includes updating the lightly trained model for a limited number of epochs to achieve an acceptable level of accuracy, and wherein the full scale training at the user device includes generating the machine-learning model using the training data and the model parameters, wherein the generating includes training a model for a large number of epochs to achieve a high level of accuracy.

18 . A non-transitory computer-readable medium storing instructions, the instructions comprising: one or more instructions that, when executed by a network element comprising one or more processors, cause the one or more processors to:

receive user device capability information from the user device;

based on the user device capability information, determine a classification of machine-learning capabilities of the user device; and

transmit, to the user device, data associated with a machine-learning model based on the classification of the machine-learning capabilities of the user device.

19 . The non-transitory computer-readable medium of claim 18 , wherein the user device capability information includes a plurality of parameters associated with the user device; and wherein determining the classification comprises:

evaluating one or more parameters of the user device from the plurality of parameters;

determining a categorical value for each of the one or more parameters; and

based on comparing the categorical value for each of the one or more parameters to a pre-determined set of minimum categorical values, assigning a classification number to the user device based on the categorical value for each of the one or more parameters.

20 . The non-transitory computer-readable medium of claim 18 , wherein transmitting data associated with a machine-learning model user device comprises:

based on the classification of the machine-learning capabilities of the user device, transmitting, to the user device, one of:

training data and model parameters for a full scale training at the user device;

a lightly trained model, a subset of the training data, and the model parameters for a lightweight training at the user device;

a general trained model, a subset of training data, and the model parameters for a specific use case-based update of the general trained model at the user device; and

a trained model for an inference at the user device.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Aug 1, 2022
From: MUHAMMAD, AWN; KITAGAWA, KOICHIRO; KESAVAN, KRISHNAKUMAR; LEE, TAEWOO
To: RAKUTEN MOBILE, INC.
Reel/Frame 060681/0453 →
Continuity (1)
Related Publication 20230351248A1 · Nov 2, 2023
References Cited (34)
US 11516311B2 · Klein · 2022 [cited by examiner]
US 11622291B2 · Pantelidou · 2023 [cited by examiner]
US 11741191B1 · Zilka · 2023 [cited by examiner]
US 11844145B2 · Elshafie · 2023 [cited by examiner]
US 11968281B2 · Klein · 2024 [cited by examiner]
US 12089291B2 · Zhu · 2024 [cited by examiner]
US 12120774B2 · Tomala · 2024 [cited by examiner]
US 12147500B2 · Zilka · 2024 [cited by examiner]
US 12245129B2 · Gulati · 2025 [cited by examiner]
US 20160253498A1 · Valencia et al. · 2016 [cited by applicant]
US 20200367051A1 · Wang et al. · 2020 [cited by applicant]
US 20220116764A1 · Pezeshki et al. · 2022 [cited by applicant]
US 20220239758A1 · Klein · 2022 [cited by examiner]
US 20220279341A1 · Tomala · 2022 [cited by examiner]
US 20220342713A1 · Shen · 2022 [cited by examiner]
US 20220383066A1 · Anicic · 2022 [cited by examiner]
US 20220400371A1 · Elshafie · 2022 [cited by examiner]
US 20230004864A1 · Wang · 2023 [cited by examiner]
US 20230044727A1 · Pantelidou · 2023 [cited by examiner]
US 20230100253A1 · Zhu · 2023 [cited by examiner]
US 20230262478A1 · Hu · 2023 [cited by examiner]
US 20230276208A1 · Ren · 2023 [cited by examiner]
US 20230276338A1 · Gulati · 2023 [cited by examiner]
US 20230344725A1 · Wang · 2023 [cited by examiner]
US 20230350978A1 · Zilka · 2023 [cited by examiner]
US 20230388817A1 · Rydén · 2023 [cited by examiner]
US 20230422117A1 · Li · 2023 [cited by examiner]
US 20250077618A1 · Zilka · 2025 [cited by examiner]
US 20250247690A1 · Ren · 2025 [cited by examiner]
Written Opinion dated Aug. 26, 2022 issued by the International Searching Authority in Application No. PCT/US2022/26646. [cited by applicant]
International Search Report dated Aug. 26, 2022 issued by the International Searching Authority in Application No. PCT/US2022/26646. [cited by applicant]
Rakuten Mobile Inc., “AI/ML Model Life Cycle Management”, 3GPP TSG RAN WG1 #109-e, R1-2205065, May 9-May 20, 2022, pp. 1-6 (6 pages total). [cited by applicant]
Samsung Electronics Co., Ltd., Discussion on AI/ML (updated), 3GPP TSG SA WG4 117-e Meeting, S4-220086, Feb. 14-24, 2022, pp. 1-9 (9 pages total). [cited by applicant]
Rakuten Mobile, “AI based enhancements for Physical Layer”, 3GPP TSG RAN Rel-18 workshop, RWS-210198, Jun. 28-Jul. 2, 2021, pp. 1-9 (9 pages total). [cited by applicant]