Base station, user equipment, network and method for machine learning related communication

CN119631380APending Publication Date: 2025-03-14CONTINENTAL AUTOMOTIVE TECHNOLOGIES GMBH
View PDF 9 Cites 0 Cited by

Patent Information

Application Number
CN202380057316.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Priority Date
2022-08-03
Filing Date
2023-07-04
Publication Date
2025-03-14

AI Technical Summary

Technical Problem

In AI/ML global model training using federated learning, when the global model convergence state is not considered, the gNB-UE signaling used to support model update information exchange may be very inefficient, resulting in increased signaling overhead.

Method used

By taking into account the global model performance state (e.g., convergence), network devices (e.g., base stations or gNBs) can enable or disable local model update reports from different numbers of user devices.

Benefits of technology

This can greatly reduce the signaling overhead in the model training stage and improve communication efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119631380A_ABST
    Figure CN119631380A_ABST
Patent Text Reader

Abstract

The present invention relates to a base station, a terminal and a method, a network and a network device, in particular a cellular mobile telecommunications network element (gNB) and / or a user equipment (UE1, UE2, UE3, UE4...), configured to enable or disable local model update reports from the user equipment (UE1, UE2, UE3, UE4...) by the network device (gNB) taking into account a global model performance state of machine learning (AI / ML).
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Federated learning across UE and ran

    US20220038349A1

  • Data transmission configuration utilizing a state indication

    US20220095146A1

  • Machine learning component update reporting in federated learning

    US20220101204A1

  • User equipment (UE) capability report for machine learning applications

    US20220116764A1

  • Wireless signaling in federated learning for machine learning components

    US20220182802A1