Method for performing federated learning in wireless communication system, and apparatus therefor

Patent metadata

Field	Value
Publication number	US-2025008449-A1
Application number	US-202118294877-A
Country	US
Kind code	A1
Filing date	Aug 3, 2021
Priority date	Aug 3, 2021
Publication date	Jan 2, 2025
Grant date	—

Abstract

Proposed are a method for performing federated learning in a wireless communication system, and an apparatus therefor. The method performed by a terminal may comprise the steps of: generating a Q-ary code including i) a restriction-based system part and ii) a parity part by coding restriction-based Q-ary information; determining, on the basis of a preset method, the number T of transmissions of the parity part of the Q-ary code; on the basis of specific channel information from among channel information between a plurality of terminals and a base station participating in the federated learning, allocating power to the system part and T parity parts; and transmitting, on the basis of the allocated power, the system part and the T parity parts to the base station.

First claim

1 . A method for performing federated learning in a wireless communication system, the method performed by a user equipment (UE) comprising:
generating a Q-ary code including i) a restriction-based system part and ii) a parity part by coding restriction-based Q-ary information; determining a number of transmissions (T) of the parity part among the Q-ary code based on a preconfigured way; allocating power to the system part and T parity parts based on specific channel information among channel information between a plurality of UEs and a base station participating in the federated learning; and transmitting the system part and the T parity parts to the base station based on the allocated power.

2 . The method of claim 1 , wherein the number of transmissions (T) is determined based on available resources.

3 . The method of claim 1 , wherein a maximum number of the transmissions of the parity part is determined based on a Q-ary related value and a restriction-based Q-ary related value.

4 . The method of claim 3 , wherein the restriction-based Q-ary related value is determined based on at least one of channel state and/or a number of the plurality of UEs.

5 . The method of claim 1 , wherein the system part is modulated based on a modulation order different from the parity part.

6 . The method of claim 1 , further comprising:
receiving the specific channel information from the base station, wherein the specific channel information is information about a channel with highest noise among channels between the plurality of UEs and the base station.

7 . A user equipment (UE) configured to perform federated learning in a wireless communication system, the UE comprising:
at least one transceiver; at least one processor functionally connected to the at least one transceiver; and at least one memory functionally connected to the at least one processor, and storing instructions for causing the at least one processor to perform operations, wherein the operations includes: generating a Q-ary code including i) a restriction-based system part and ii) a parity part by coding restriction-based Q-ary information; determining a number of transmissions (T) of the parity part among the Q-ary code based on a preconfigured way; allocating power to the system part and T parity parts based on specific channel information among channel information between a plurality of UEs and a base station participating in the federated learning; and transmitting the system part and the T parity parts to the base station based on the allocated power.

8 . A method for performing federated learning in a wireless communication system, the method performed by a base station comprising:
receiving a system part and T parity parts from a user equipment (UE) based on allocated power, wherein the system part and parity parts are generated by coding restriction-based Q-ary information, wherein a number of transmissions (T) of the parity part is determined based on a preconfigured way, and wherein the allocated power of the system part and T parity parts is determined based on specific channel information among channel information between a plurality of UEs and a base station participating in the federated learning.

9 . The method of claim 8 , wherein the number of transmissions (T) is determined based on available resources.

10 . The method of claim 8 , wherein a maximum number of the transmissions of the parity part is determined based on a Q-ary related value and a restriction-based Q-ary related value.

11 . The method of claim 10 , wherein the restriction-based Q-ary related value is determined based on at least one of channel state and/or a number of the plurality of UEs.

12 . The method of claim 8 , wherein the system part is modulated based on a modulation order different from the parity part.

13 . The method of claim 8 , further comprising:
transmitting the specific channel information to the UE, wherein the specific channel information is information about a channel with highest noise among channels between the plurality of UEs and the base station.

14 - 16 . (canceled)

Assignees

• Lg Electronics Inc

Inventors

• Jeon Kijun
• Lee Sangrim
• Jung Ikjoo
• Lee Hojae
• Kim Yeongjun
• Lee Taehyun

Classifications

• G06N3/084

  Backpropagation, e.g. using gradient descent · CPC title

• G06N3/0464

  Convolutional networks [CNN, ConvNet] · CPC title

• G06N3/0455

  Auto-encoder networks; Encoder-decoder networks · CPC title

• G06N3/044 Primary

  Recurrent networks, e.g. Hopfield networks · CPC title

• H04W52/243

  taking into account interferences · CPC title