Performance monitoring of a two-sided artificial intelligence / machine learning model at the user equipment side

EP4758907A1Pending Publication Date: 2026-06-17TELEFONAKTIEBOLAGET LM ERICSSON (PUBL)
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
EP · EP
Patent Type
Applications
Current Assignee / Owner
TELEFONAKTIEBOLAGET LM ERICSSON (PUBL)
Filing Date
2024-08-09
Publication Date
2026-06-17

AI Technical Summary

Technical Problem

Existing methods for monitoring the performance of two-sided AI/ML models at the user equipment (UE) side often result in increased signaling overhead and other drawbacks, such as exposing proprietary model information and requiring significant data transmission.

Method used

A method where the UE measures reference signal resources, performs model inference, estimates intermediate Key Performance Indicators (KPIs), and reports these estimates to the network node, using a one-sided AI/ML model to predict intermediate KPI range indicators, thereby reducing signaling overhead and maintaining model privacy.

Benefits of technology

This approach enables efficient monitoring and reporting of two-sided AI/ML model performance with reduced signaling overhead, while protecting proprietary model information, thus improving operational efficiency and security.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure SE2024050724_20022025_PF_FP_ABST
    Figure SE2024050724_20022025_PF_FP_ABST
Patent Text Reader

Abstract

A method performed by a user equipment is provided The method comprises measuring at least one reference signal resource indicated by a reference signal configuration associated with a two-sided AI / ML model. The two-sided AI / ML model comprises a UE part operated by the UE and a network part operated by a network node. The method further comprises performing model inference based on the measuring of the at least one reference signal resource using the UE part of the two-sided AI / ML model. The method further comprises estimating at least one intermediate KPI of the two-sided AI / ML model. The method further comprises reporting the estimated at least one intermediate KPI to the network node.
Need to check novelty before this filing date? Find Prior Art