公开(公告)号：US20210241173A1

公开(公告)日：2021-08-05

申请号：US17008218

申请日：2020-08-31

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Hang Li ,  Xi Chen ,  Di Wu ,  Xue Liu

IPC: G06N20/00 ,  H04W4/02 ,  G01P13/00 ,  G01S5/02 ,  G06N5/04

Abstract: A method of location determination with a WiFi transceiver and an AI model includes jointly training, based on various losses: a feature extractor, a location classifier, and a domain classifier. The domain classifier may include a first domain classifier and a second domain classifier. The losses used for training tend to cause feature data from the feature extractor to cluster even if a physical object in an environment has moved after training is completed. Then, the location classifier is able to accurately estimate the position of, for example, a person in a room, even if a door or window has changed from open to close or close to open between the time of training and the time of estimating the person's position.

公开(公告)号：US12052584B2

公开(公告)日：2024-07-30

申请号：US18199666

申请日：2023-05-19

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Xi Chen ,  Ju Wang ,  Hang Li ,  Yi Tian Xu ,  Di Wu ,  Xue Liu ,  Gregory Lewis Dudek ,  Taeseop Lee ,  Intaik Park

IPC: H04W16/18 ,  H04W16/22

CPC classification number: H04W16/22 ,  H04W16/18

Abstract: Transfer learning based on prediction determines a similarity between a source base station and a target base station. Importance of parameters is determined and training is adjusted to respect the importance of parameters. A lack of historical data is compensated by selecting a base station as source base station which has a larger amount of historical data.

公开(公告)号：US20230353659A1

公开(公告)日：2023-11-02

申请号：US18351201

申请日：2023-07-12

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Jikun KANG ,  Xi Chen ,  Chengming Hu ,  Ju Wang ,  Gregory Lewis Dudek ,  Xue Liu

IPC: H04L67/5682 ,  H04L67/1004 ,  H04L41/16

CPC classification number: H04L67/5682 ,  H04L67/1004 ,  H04L41/16

Abstract: A server may be provided to obtain a load balancing artificial intelligence (AI) model for a plurality of base stations in a communication system. The server may obtain teacher models based on traffic data sets collected from the base stations, respectively; perform a policy rehearsal process including obtaining student models based on knowledge distillation from the teacher models, obtaining an ensemble student model by ensembling the student models, and obtaining a policy model by interacting with the ensemble student mode; provide the policy model to each of the base stations for a policy evaluation of the policy model; and based on a training continue signal being received from at least one of the base stations as a result of the policy evaluation, update the ensemble student model and the policy model by performing the policy rehearsal process on the student models.

公开(公告)号：US11751115B2

公开(公告)日：2023-09-05

申请号：US17363918

申请日：2021-06-30

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Jikun Kang ,  Xi Chen ,  Di Wu ,  Yi Tian Xu ,  Xue Liu ,  Gregory Lewis Dudek ,  Taeseop Lee ,  Intaik Park

IPC: H04W36/22 ,  G06N3/08 ,  H04W24/02 ,  G06N3/044 ,  H04W28/086

CPC classification number: H04W36/22 ,  G06N3/044 ,  G06N3/08 ,  H04W24/02 ,  H04W28/0861

Abstract: Hybrid use of dual policies is provided to improve a communication system. In a multiple access scenario, when an inactive user equipment (UE) transitions to an active state, it may be become a burden to a radio cell on which it was previously camping. In some embodiments, hybrid load balancing is provided using a hierarchical machine learning paradigm based on reinforcement learning in which an LSTM generates a goal for one policy influencing cell reselection so that another policy influencing handover over active UEs can be assisted. The communication system as influenced by the policies is modeled as a Markov decision process (MDP). The policies controlling the active UEs and inactive UEs are coupled, and measureable system characteristics are improved. In some embodiments, policy actions depend at least in part on energy saving.

公开(公告)号：US11461699B2

公开(公告)日：2022-10-04

申请号：US17008218

申请日：2020-08-31

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Hang Li ,  Xi Chen ,  Di Wu ,  Xue Liu

IPC: G06N20/00 ,  H04W4/02 ,  G01S5/02 ,  G06N5/04 ,  G01P13/00 ,  H04W84/12

Abstract: A method of location determination with a WiFi transceiver and an AI model includes jointly training, based on various losses: a feature extractor, a location classifier, and a domain classifier. The domain classifier may include a first domain classifier and a second domain classifier. The losses used for training tend to cause feature data from the feature extractor to cluster even if a physical object in an environment has moved after training is completed. Then, the location classifier is able to accurately estimate the position of, for example, a person in a room, even if a door or window has changed from open to close or close to open between the time of training and the time of estimating the person's position.

公开(公告)号：US20250168255A1

公开(公告)日：2025-05-22

申请号：US19028822

申请日：2025-01-17

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Jikun Kang ,  Xi Chen ,  Chengming Hu ,  Ju Wang ,  Gregory Lewis Dudek ,  Xue Liu

IPC: H04L67/5682 ,  H04L41/16 ,  H04L67/1004

Abstract: A server may be provided to obtain a load balancing artificial intelligence (AI) model for a plurality of base stations in a communication system. The server may obtain teacher models based on traffic data sets collected from the base stations, respectively; perform a policy rehearsal process including obtaining student models based on knowledge distillation from the teacher models, obtaining an ensemble student model by ensembling the student models, and obtaining a policy model by interacting with the ensemble student mode; provide the policy model to each of the base stations for a policy evaluation of the policy model; and based on a training continue signal being received from at least one of the base stations as a result of the policy evaluation, update the ensemble student model and the policy model by performing the policy rehearsal process on the student models.

公开(公告)号：US12238190B2

公开(公告)日：2025-02-25

申请号：US18351201

申请日：2023-07-12

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Jikun Kang ,  Xi Chen ,  Chengming Hu ,  Ju Wang ,  Gregory Lewis Dudek ,  Xue Liu

IPC: H04L67/5682 ,  H04L41/16 ,  H04L67/1004

Abstract: A server may be provided to obtain a load balancing artificial intelligence (AI) model for a plurality of base stations in a communication system. The server may obtain teacher models based on traffic data sets collected from the base stations, respectively; perform a policy rehearsal process including obtaining student models based on knowledge distillation from the teacher models, obtaining an ensemble student model by ensembling the student models, and obtaining a policy model by interacting with the ensemble student mode; provide the policy model to each of the base stations for a policy evaluation of the policy model; and based on a training continue signal being received from at least one of the base stations as a result of the policy evaluation, update the ensemble student model and the policy model by performing the policy rehearsal process on the student models.

公开(公告)号：US11825371B2

公开(公告)日：2023-11-21

申请号：US17334018

申请日：2021-05-28

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Di Wu ,  Jikun Kang ,  Yi Tian Xu ,  Jimmy Li ,  Michael Jenkin ,  Xue Liu ,  Xi Chen ,  Gregory Lewis Dudek ,  Intaik Park ,  Taeseop Lee

IPC: H04W36/22 ,  H04W36/08 ,  G06N3/08 ,  G06N20/00 ,  H04W48/06 ,  G06N3/045

CPC classification number: H04W36/22 ,  G06N3/045 ,  G06N3/08 ,  G06N20/00 ,  H04W36/08 ,  H04W48/06

Abstract: An apparatus distributing communication load over a plurality of communication cells may select action centers from random cell reselection values, based on a standard deviation of an internet protocol (IP) throughout over the plurality of communication cells; input a first vector indicating a communication state of a communication system and a second vector indicating the standard deviation of the IP throughout of the plurality of communication cells, to a neural network to output a sum of the action centers and offsets as cell reselection parameters; and transmit the cell reselection parameters to the communication system to enable a base station of the communication system to perform a cell reselection based on the cell reselection parameters.

公开(公告)号：US20230247509A1

公开(公告)日：2023-08-03

申请号：US18133845

申请日：2023-04-12

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Jikun KANG ,  Xi Chen ,  Di Wu ,  Yi Tian Xu ,  Xue Liu ,  Gregory Lewis Dudek ,  Taeseop Lee ,  Intaik Park

IPC: H04W36/22 ,  H04W28/086 ,  G06N3/08 ,  H04W24/02 ,  G06N3/044

CPC classification number: H04W36/22 ,  H04W28/0808 ,  G06N3/08 ,  H04W24/02 ,  G06N3/044

Abstract: Hybrid use of dual policies is provided to improve a communication system. In a multiple access scenario, when an inactive user equipment (UE) transitions to an active state, it may be become a burden to a radio cell on which it was previously camping. In some embodiments, hybrid load balancing is provided using a hierarchical machine learning paradigm based on reinforcement learning in which an LSTM generates a goal for one policy influencing cell reselection so that another policy influencing handover over active UEs can be assisted. The communication system as influenced by the policies is modeled as a Markov decision process (MDP). The policies controlling the active UEs and inactive UEs are coupled, and measureable system characteristics are improved. In some embodiments, policy actions depend at least in part on energy saving.

公开(公告)号：US11696153B2

公开(公告)日：2023-07-04

申请号：US17391708

申请日：2021-08-02

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Xi Chen ,  Ju Wang ,  Hang Li ,  Yi Tian Xu ,  Di Wu ,  Xue Liu ,  Gregory Lewis Dudek ,  Taeseop Lee ,  Intaik Park

IPC: H04W16/18 ,  H04W16/22

CPC classification number: H04W16/22 ,  H04W16/18

Abstract: Transfer learning based on prediction determines a similarity between a source base station and a target base station. Importance of parameters is determined and training is adjusted to respect the importance of parameters. A lack of historical data is compensated by selecting a base station as source base station which has a larger amount of historical data.