TECHNIQUES FOR MODIFYING MACHINE LEARNING MODELS USING IMPORTANCE WEIGHTS

    公开(公告)号:US20250106653A1

    公开(公告)日:2025-03-27

    申请号:US18818273

    申请日:2024-08-28

    Abstract: Methods, systems, and devices for wireless communication are described. A network node may calculate a set of weights for a first set of data associated with training a machine learning (ML) model in accordance with a first set of operating conditions for maintaining a communication link. Each weight of the set of weights may be associated with a respective datum of the first set of data and may be based on a probability that the respective datum is included in a second set of data. The second set of data may be associated with obtaining predictions using the machine learning model in accordance with a second set of operating conditions for maintaining the communication link. The network node may identify an event associated with the ML model and may output information indicative of the set of weights based on the event.

    SOFT INTERFERENCE PREDICTION IN A WIRELESS COMMUNICATIONS SYSTEM

    公开(公告)号:US20250071791A1

    公开(公告)日:2025-02-27

    申请号:US18941990

    申请日:2024-11-08

    Abstract: Certain aspects of the present disclosure provide techniques for wireless communications by a user equipment (UE), including receiving, from a network entity, one or more parameters for performing interference prediction at the UE; predicting, for each of at least one communications resource, a predicted interference at the UE, wherein the predicted interference comprises a plurality of probability values, each probability value of the plurality of probability values being associated with a different class of a plurality of classes, each class of the plurality of classes associated with a corresponding range of interference power; and sending, to the network entity, a report based on the predicted interference.

    REPETITION REQUEST FOR COVERAGE ENHANCEMENT

    公开(公告)号:US20250007654A1

    公开(公告)日:2025-01-02

    申请号:US18292835

    申请日:2021-09-10

    Abstract: Methods, apparatuses, and computer-readable medium for reducing signaling overhead for a RedCap UE are provided. An example method may include receiving, from a base station, a grant associated with a set of random access channel (RACH) resources, a first part of the set of RACH resources for a reduced capability report including a set of overlapping resources that overlaps with a second part of the set of RACH resources for a physical uplink shared channel (PUSCH) repetition request. The example method may further include transmitting, in the set of overlapping resources to the base station, one or more repetition requests or one or more repetition number indications.

    BEAM BLOCKAGE PREDICTION AND REPORTING

    公开(公告)号:US20240413870A1

    公开(公告)日:2024-12-12

    申请号:US18699860

    申请日:2021-12-22

    Abstract: Methods, apparatuses, and computer readable medium for wireless communication are provided. An example method may include receiving, from a base station, at least one channel status information (CSI) report setting associated with CSI resource setting that configures a set of CSI reference signals (CSI-RS resources), at least one CSI report setting including at least one quantity change rate associated with one or more predicted quantities over at least one future time window or one or more measured quantities over at least one past time window, one or more predicted quantities or one or more measured quantities being associated with the set of CSI-RS resources. The example UE may further include transmitting, to the base station based on at least one CSI report setting, the quantity change rate or reliability information message based on one or more predicted quantities over at least one future time window.

    EFFICIENT CODEBOOK-BASED INTERFERENCE PREDICTION REPORTING

    公开(公告)号:US20240340125A1

    公开(公告)日:2024-10-10

    申请号:US18295569

    申请日:2023-04-04

    CPC classification number: H04L5/0032 H04L5/0096 H04W72/02 H04W72/21

    Abstract: Methods, systems, and devices for wireless communications at a user equipment (UE) are described. The UE may receive a control message indicating a set of multiple resources for which the UE may predict interference information. The UE may select a codebook from a set of multiple codebooks to compress the interference information for the set of multiple resources that follows the control message. The UE may select the codebook based on the set of multiple resources associated with the interference information The UE may transmit a report indicating the interference information predictions to a network entity. That is, the report may include the interference information that may be compressed using the selected codebook. The network entity may receive the report and decompress the interference information to retrieve the predicted interference information for a wireless communications system.

    NON-COHERENT MODULATION FOR FEDERATED LEARNING

    公开(公告)号:US20240322956A1

    公开(公告)日:2024-09-26

    申请号:US18186920

    申请日:2023-03-20

    Abstract: A UE may identify, in at least one round of a federated learning procedure, at least one gradient update based on local data and a local copy of a machine learning model associated with the federated learning procedure. The UE may transmit, in the at least one round of the federated learning procedure, for a network node, an indication of the at least one gradient update based on a sign of the at least one gradient update and a non-coherent orthogonal modulation scheme. The network node may identify, in at least one round of the federated learning procedure, an accumulated sign of the at least one gradient update based on a first received power associated with the at least one first resource and a second received power associated with the at least one second resource.

Patent Agency Ranking