Communication method and apparatus

By selecting appropriate clients to participate in training based on terminal information in distributed learning, the problem of poor learning results is solved, achieving efficient utilization of terminal resources and low-power distributed learning.

WO2026124306A1 Publication Date: 2026-06-18HUAWEI TECH CO LTD
0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
HUAWEI TECH CO LTD
Filing Date
2025-12-02
Publication Date
2026-06-18

AI Technical Summary

⚠Technical Problem

In distributed learning, failure to properly select clients to participate in training can lead to poor learning outcomes or the need to increase terminal power consumption to maintain the results.

⚗Method used

The first node selects terminals to participate in distributed learning based on their connection status, energy saving, and capability information, ensuring that the terminals are in RRC connection state, the transmission time interval is appropriate, the transmission rate meets the requirements, and the number reaches the threshold, thereby improving the learning effect and reducing power consumption.

🎯Benefits of technology

This improves the effectiveness of distributed learning while avoiding unnecessary increases in terminal power consumption and optimizing terminal resource utilization.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 1
    Figure 1
  • Figure 2
    Figure 2
  • Figure 3
    Figure 3
Patent Text Reader

Abstract

A communication method and apparatus, which are used for rationally selecting a client participating in training, so as to improve a distributed learning effect without increasing the overall power consumption of a terminal. Taking the method being applied to a first node as an example, in the method, the first node can obtain one or more of the following pieces of information of a terminal: connection state information, energy saving information, and capability information. The first node can further generate first response information on the basis of the one or more pieces of information, wherein the first response information is used for indicating one or more first terminals participating in distributed learning and / or regions where the first terminals are located.
Need to check novelty before this filing date? Find Prior Art