公开(公告)号：US20240322941A1

公开(公告)日：2024-09-26

申请号：US18573914

申请日：2021-06-23

Applicant: LG Electronics Inc.

Inventor： Kijun JEON ,  Sangrim LEE ,  Hojae LEE ,  Yeongjun KIM ,  Taehyun LEE

IPC: H04L1/00

CPC classification number: H04L1/0041 ,  H04L1/0061

Abstract: The present disclosure provides a method for performing federated learning by one user equipment (UE) with a plurality of user equipments (UEs) in a wireless communication system. More specifically, the method performed by one UE comprises receiving, from a base station, restriction information related to an order restriction for an information field for generating a local parameter transmitted by the one UE to perform the federated learning; performing encoding on the local parameter generated from an order restricted information field based on the restriction information, wherein the encoded local parameter consists of a systematic bit part and a parity part; performing modulation on the encoded local parameter, wherein the modulation is performed in different ways for each of the systematic bit part and the parity part; and transmitting, to the base station, a signal including the modulated local parameters.

公开(公告)号：US20240129755A1

公开(公告)日：2024-04-18

申请号：US18267666

申请日：2020-12-31

Applicant: LG Electronics Inc.

Inventor： Kijun JEON ,  Sangrim LEE ,  Sungjin KIM ,  Hojae LEE ,  Yeongjun KIM ,  Taehyun LEE

IPC: H04W24/02 ,  H04W24/10 ,  H04W52/50 ,  H04W74/00 ,  H04W74/0833

CPC classification number: H04W24/02 ,  H04W24/10 ,  H04W52/50 ,  H04W74/002 ,  H04W74/0833

Abstract: In the present specification, provided are an apparatus and a method for obtaining gradient information, performed by a terminal, in a wireless communication system, wherein the terminal transmits a random access (RA) preamble to a base station, receives a random access response (RAR) from the base station in response to the RA preamble, performs a radio resource control (RRC) connection procedure with the base station, establishes a digital transmission link on the basis of the RRC connection procedure with the base station, and establishes an airComp link that is a link for federated learning with the base station, wherein the terminal acquires the gradient information corresponding to data, and the terminal obtains first information and second information from the gradient information on the basis of data splitting modulation, wherein the first information is information having a lower priority among the information obtained on the basis of the gradient information, the second information is information having a high priority among the information obtained on the basis of the gradient information, the terminal transmits the first information to the base station on a first area, and the terminal transmits the second information to the base station on a second area, wherein the first area is an area distinct from the second area in terms of a frequency domain and a time domain, the first area is a resource area used on the airComp link, and the second region is a resource area used on a digital transmission link.

公开(公告)号：US20230318662A1

公开(公告)日：2023-10-05

申请号：US18042107

申请日：2020-08-20

Applicant: LG ELECTRONICS INC.

Inventor： Kijun JEON ,  Sangrim LEE ,  Hojae LEE ,  Ikjoo JUNG ,  Taehyun LEE

IPC: H04B7/0404 ,  G06N3/045 ,  H04B7/0452

CPC classification number: H04B7/0404 ,  G06N3/045 ,  H04B7/0452

Abstract: Disclosed is a method by which a terminal controls the calculations of a deep neural network in a wireless communication system. A method according to one embodiment of the present disclosure receives a downlink from a base station of a wireless communication system by using a multi input multi output (MIMO) operation, pre-processes the downlink on the basis of a result of calculations of a deep neural network in a terminal, acquires the number of a plurality of transmission antennas connected to the terminal, acquires the number of a plurality of reception antennas connected to the terminal, and forms a plurality of overlapping neural networks overlapping in the deep neural network, on the basis of a preset number of reference antennas, the number of transmission antennas, and the number of reception antennas. The terminal of the present disclosure can be linked to an artificial intelligence module, a drone (unmanned aerial vehicle (UAV)), a robot, an augmented reality (AR) device, a virtual reality (VR) device, a device related to 6G services, and the like.

公开(公告)号：US20240073732A1

公开(公告)日：2024-02-29

申请号：US18033876

申请日：2021-10-28

Applicant: LG ELECTRONICS INC.

Inventor： Ikjoo JUNG ,  Sangrim LEE ,  Jayeong KIM ,  Taehyun LEE

IPC: H04W28/02 ,  G06N3/09 ,  H04L5/00 ,  H04W72/21

CPC classification number: H04W28/0236 ,  G06N3/09 ,  H04L5/0057 ,  H04W72/21

Abstract: In order to provide device and method for adjusting split point in wireless communication system, a method of operating a terminal in a wireless communication system may comprise requesting a base station for uplink channel quality information, receiving, from the base station, the uplink channel quality information, determining a first split point based on the channel quality information, performing first split inference based on the determined first split point and transmitting, to the base station, first split inference result information.

公开(公告)号：US20250008449A1

公开(公告)日：2025-01-02

申请号：US18294877

申请日：2021-08-03

Applicant: LG Electronics Inc.

Inventor： Kijun JEON ,  Sangrim LEE ,  Ikjoo JUNG ,  Hojae LEE ,  Yeongjun KIM ,  Taehyun LEE

IPC: H04W52/32 ,  H04L1/00 ,  H04L41/16

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.

公开(公告)号：US20240394602A1

公开(公告)日：2024-11-28

申请号：US18687698

申请日：2021-09-01

Applicant: LG ELECTRONICS INC.

Inventor： Taehyun LEE ,  Sangrim LEE ,  Ikjoo JUNG ,  Hojae LEE ,  Yeongjun KIM ,  Kijun JEON

IPC: G06N20/00

Abstract: The present specification provides a method by which a plurality of terminals perform federated learning in a wireless communication system. More particularly, the method performed by one terminal comprises the steps of: receiving, from a base station, a first downlink signal for requesting information regarding learning data for the federated learning, which is used by the one terminal; transmitting, to the base station, the information regarding the learning data on the basis of a type of learning data, the information regarding the learning data being information related to the distribution of the learning data used by the one terminal; receiving, from the base station, a second downlink signal including parameter information regarding a parameter related to a configuration for performing the federated learning, which has been determined on the basis of the learning data; and performing the federated learning on the basis of the information regarding the parameter.

公开(公告)号：US20240223407A1

公开(公告)日：2024-07-04

申请号：US18563196

申请日：2021-05-21

Applicant: LG Electronics Inc.

Inventor： Yeongjun KIM ,  Sangrim LEE ,  Kijun JEON ,  Taehyun LEE

IPC: H04L25/02 ,  G06N3/098 ,  H04W72/23

CPC classification number: H04L25/024 ,  G06N3/098 ,  H04L25/0204 ,  H04W72/23

Abstract: The present disclosure provides a method of performing, by a plurality of user equipments (UEs), a federated learning in a wireless communication system. More specifically, the method performed by one UE of the plurality of UEs comprises receiving, from a base station (BS), control information related to a scheduling of a resource on which the one UE repeatedly transmits a local parameter of the one UE, wherein a global parameter for the federated learning is updated based on respective local parameters of the plurality of UEs: transmitting, to the BS, a first signal including the local parameter of the one UE on a first resource scheduled based on the control information; and transmitting, to the BS, a second signal including the local parameter of the one UE on a second resource that is scheduled based on the control information and is different from the first resource, wherein the first signal and the second signal are transmitted based on the first signal and the second signal being in a complex conjugate relationship.

公开(公告)号：US20250023609A1

公开(公告)日：2025-01-16

申请号：US18711460

申请日：2022-11-07

Applicant: LG ELECTRONICS INC.

Inventor： Kijun JEON ,  Sangrim LEE ,  Ikjoo JUNG ,  Hojae LEE ,  Taehyun LEE

IPC: H04B7/06 ,  H04L5/00 ,  H04L41/16 ,  H04W52/36

Abstract: The present specification provides a method by which a terminal performs federated learning with a plurality of terminals in a wireless communication system. More specifically, the method performed by one terminal comprises the steps of: receiving, from a server, a channel state information reference signal (CSI-RS); transmitting, to the server, channel state information (CSI) calculated on the basis of the CSI-RS; receiving, from the server, (i) information about a global parameter for the federated learning and (ii) compression state information for determining a weight compression method of the one terminal on the basis of channel state information of each of channels between the server and the plurality of terminals; determining a weight compression scheme based on (i) a difference value between the global parameter and a global parameter received before receiving the global parameter and (ii) the compression state information; and transmitting, to the server, an updated local parameter on the basis of the determined weight compression scheme.

公开(公告)号：US20250016619A1

公开(公告)日：2025-01-09

申请号：US18712453

申请日：2022-11-07

Applicant: LG ELECTRONICS INC.

Inventor： Kijun JEON ,  Sangrim LEE ,  Ikjoo JUNG ,  Hojae LEE ,  Taehyun LEE

IPC: H04W28/06 ,  G06N3/098

Abstract: The present disclosure provides a method for one user equipment (UE) to perform federated learning with a plurality of UEs in a wireless communication system. More specifically, the method performed by the one UE comprises receiving, from a server, a channel state information reference signal (CSI-RS); transmitting, to the server, channel state information (CSI) calculated based on the CSI-RS; receiving, from the server, compression state information for determining a weight compression method of the one UE based on (i) information on a global parameter for the federated learning and (ii) channel state information of each of channels between the server and the plurality of UEs; determining the weight compression method based on (i) a difference between the global parameter and a global parameter received before a reception of the global parameter and (ii) the compression state information; and transmitting, to the server, a local parameter updated based on the determined weight compression method.

公开(公告)号：US20240414746A1

公开(公告)日：2024-12-12

申请号：US18702691

申请日：2022-10-17

Applicant: LG ELECTRONICS INC.

Inventor： Taehyun LEE ,  Sangrim LEE ,  Kijun JEON ,  Ikjoo JUNG ,  Yeongjun KIM

IPC: H04W72/51 ,  H04W56/00 ,  H04W72/04

Abstract: The present disclosure provides a device and method for performing, based on channel information, a device grouping for federated learning based AirCOMP of a non-IID data environment in a communication system. The present disclosure also provides a device and method for performing effective federated learning in a non-IID environment including multiple devices. The present disclosure also provides a device and method for performing a device grouping in consideration of channel environment factors in order to apply AirComp based federated learning to a real communication environment.