MODEL TRAINING METHOD AND RELATED APPARATUS
    111.
    发明公开

    公开(公告)号:US20240152766A1

    公开(公告)日:2024-05-09

    申请号:US18405019

    申请日:2024-01-05

    CPC classification number: G06N3/096

    Abstract: A model training method and a related apparatus to help improve a convergence speed of model training and improve end-to-end communication quality. The method includes: a first communication apparatus sends first data to a second communication apparatus through a channel, where the first data is an output result of the first machine learning model. The second communication apparatus receives second data through a channel, inputs the second data into a second machine learning model to obtain third data; determines a first loss function based on the third data and the first training data; and sends the first loss function to the first communication apparatus through a feedback channel.

    ACCESS METHOD AND COMMUNICATION APPARATUS
    112.
    发明公开

    公开(公告)号:US20240107427A1

    公开(公告)日:2024-03-28

    申请号:US18519911

    申请日:2023-11-27

    CPC classification number: H04W48/10 H04L5/0048 H04W48/16

    Abstract: This application provides an access method and a communication apparatus, applied to the communication field, and in particular, to the field of short-range communication. The access method includes: sending first broadcast information based on a target access mode, where the first broadcast information indicates that a broadcast type is connectable; and if a first access request from a first device is received within first preset duration, sending first response information for the first access request, where the target access mode is an access mode in an access mode set, and the access mode set includes at least two access modes. Through configuration of the at least two access modes, access modes can be selected in different scenarios, to shorten time for subsequently establishing a connection between the first device and a second device, reduce power consumption of the first device or the second device.

    DISTRIBUTED LEARNING METHOD AND APPARATUS
    113.
    发明公开

    公开(公告)号:US20240054324A1

    公开(公告)日:2024-02-15

    申请号:US18486807

    申请日:2023-10-13

    CPC classification number: G06N3/045 G06N3/082

    Abstract: A distributed learning method and apparatus for combining wireless communication with distributed learning to save resources, and improve performance of distributed learning in a wireless environment. A first node processes first data using a first data model to obtain first intermediate data. The first node sends the first intermediate data to a second node through a first channel. The first channel is updated based on error information of second intermediate data, information about the first channel, and the first intermediate data. The second intermediate data is a result of transmitting the first intermediate data to the second node through the first channel. The first channel is a channel between the first node and the second node.

    GRADIENT TRANSMISSION METHOD AND RELATED APPARATUS

    公开(公告)号:US20240049188A1

    公开(公告)日:2024-02-08

    申请号:US18486482

    申请日:2023-10-13

    CPC classification number: H04W72/04 H04L41/16

    Abstract: A first communication apparatus receives training data, and determines a first intermediate gradient based on the training data. The first intermediate gradient is used to update a parameter of a first neural network located in a second communication apparatus. The first communication apparatus maps the first intermediate gradient to an air interface resource to generate a first gradient signal, and sends the first gradient signal to the second communication apparatus. The first gradient signal includes one or more first gradient symbols, and each of the first gradient symbols is corresponding to one or more gradient values.

    INFORMATION GENERATION METHOD AND RELATED APPARATUS

    公开(公告)号:US20230336382A1

    公开(公告)日:2023-10-19

    申请号:US18339474

    申请日:2023-06-22

    CPC classification number: H04L25/0254 G06N3/045 G06N3/08

    Abstract: The method is provided, including: A second device receives a first message and a third message, and sends a second message to a first device. The first message indicates all or a part of a first generator, the third message indicates all or a part of a third generator, an input supported by the first generator includes first information of a first type, an input supported by the third generator includes fourth information of the first type, and the first generator and the third generator are configured to train a neural network corresponding to a second generator; and the second message indicates all or a part of the second generator, and an input supported by the second generator includes the first information and the fourth information. Information collected in a real scenario may be used to train a generation model, to make a communication network more intelligent and adaptive.

    DATA PACKET DECOMPRESSION METHOD, ELECTRONIC DEVICE, AND NETWORK SIDE DEVICE

    公开(公告)号:US20230308944A1

    公开(公告)日:2023-09-28

    申请号:US18041178

    申请日:2021-08-09

    CPC classification number: H04W28/06 H04W28/0278 H04W28/04 H04W80/02

    Abstract: Embodiments of this application provide a data packet decompression method. In the method, the network side device receives a first PDCP data packet, decompresses the first PDCP data packet, and sends a first buffer data identifier to the electronic device if the first PDCP data packet fails to be decompressed, where the first buffer data identifier corresponds to buffer data used for decompressing the first PDCP data packet; the electronic device receives the first buffer data identifier, and sends first buffer data corresponding to the first buffer data identifier to the network side device, where the first buffer data is buffer data used for generating the first PDCP data packet; and the network side device receives the first buffer data, and decompresses the first PDCP data packet by using the first buffer data.

    COMMUNICATION METHOD AND APPARATUS
    117.
    发明公开

    公开(公告)号:US20230198826A1

    公开(公告)日:2023-06-22

    申请号:US18173221

    申请日:2023-02-23

    CPC classification number: H04L27/3472 H04L43/08

    Abstract: A communication method and apparatus are configured to design a constellation diagram that matches a communication scenario. The constellation diagram can adapt to different communication scenarios and meet different communication performance requirements, thereby improving network performance. In the communication method, a first communication apparatus sends a constellation diagram design parameter, where the constellation diagram design parameter includes a communication scenario design parameter and a communication apparatus design parameter. The first communication apparatus receives information about a second constellation diagram, where the second constellation diagram corresponds to the constellation diagram design parameter. The first communication apparatus performs communication by using the second constellation diagram.

    CHANNEL ENCODING AND DECODING METHOD AND COMMUNICATION APPARATUS

    公开(公告)号:US20230123083A1

    公开(公告)日:2023-04-20

    申请号:US18067244

    申请日:2022-12-16

    Abstract: This application provides a channel encoding method and a communication apparatus. A second communication apparatus obtains a first parameter of a first communication apparatus, where the first parameter includes a parameter related to channel coding and decoding and a reinforcement learning training parameter. The second communication apparatus determines, based on the first parameter, first code construction information for constructing a coded bit sequence based on an information bit sequence during channel encoding; and after sending the first code construction information to the first communication apparatus, performs channel encoding and decoding on communication data between the first communication apparatus and the second communication apparatus by using the first code construction information to improve channel encoding performance and further improve communication reliability.

    COMMUNICATION METHOD AND APPARATUS
    119.
    发明申请

    公开(公告)号:US20230097297A1

    公开(公告)日:2023-03-30

    申请号:US17801184

    申请日:2020-10-20

    Abstract: This application relates to a communication method and apparatus. A first device determines to process a first radio bearer in a first processing manner. The first processing manner activates or deactivates a data packet compression function of the first radio bearer. Then, the first device sends first information, where the first information is used to indicate the first processing manner. The first device can send the first information to indicate to activate or deactivate the data packet compression function, which means that the data packet compression function can be deactivated.

    SCHEDULING REQUEST CONFIGURATION METHOD AND SENDING METHOD, AND CORRESPONDING APPARATUS

    公开(公告)号:US20230085148A1

    公开(公告)日:2023-03-16

    申请号:US17939624

    申请日:2022-09-07

    Abstract: A scheduling request (SR) configuration method and sending method, and a corresponding apparatus are provided, to resolve a prior-art problem of a scheduling request configuration method and sending method, and a corresponding apparatus. The scheduling request configuration method includes: receiving, by a user equipment (UE), control signaling sent by a network device, where the control signaling is used to configure, for the UE, an SR configuration associated with at least one logical channel and at least two SR resource configurations associated with the SR configuration, and the SR resource configurations indicate physical resources used to signal an SR associated with the at least one logical channel; and signaling, by the UE based on a physical resource indicated by one of the at least two SR resource configurations, an SR associated with one of the at least one logical channel.

Patent Agency Ranking