Input selection aware monitoring for enhanced machine learning based positioning

By assessing and monitoring input matching between training and inference phases, the method ensures consistent input selection for machine learning-based positioning, improving accuracy in dynamic TRP environments.

WO2025233560A1 Publication Date: 2025-11-13NOKIA TECHNOLOGIES OY
2 Cites 1 Cited by

Patent Information

Application Number
PCT/FI2025/050060
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2024-05-10
Filing Date
2025-02-12
Publication Date
2025-11-13

AI Technical Summary

Technical Problem

Existing communication systems face challenges in ensuring consistent input matching between training and inference phases in machine learning-based positioning, leading to varying positioning accuracy due to differences in transmission-reception-point (TRP) selection and order, particularly in non-line-of-sight scenarios.

Method used

Implementing a method to assess and monitor the matching of inputs used during training and inference phases, using threshold matching factors and channel characteristics to adjust machine learning model switching, ensuring consistent input selection for improved positioning accuracy.

Benefits of technology

Enhances positioning accuracy by ensuring consistent input selection, thereby meeting quality of service requirements in dynamic TRP environments.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

A method for a first apparatus, the method comprising: obtaining (713) at least one matching factor, wherein the at least one matching factor indicates a degree matching between inputs used during a machine learning training phase and inputs used during a machine learning inference phase; obtaining at least one threshold matching factor from a second apparatus; comparing (715) the at least one threshold matching factor and the at least one matching factor; performing at least a machine learning model switching based on the monitoring decision request from the second apparatus, the monitoring decision request based at least on: the obtained at least one matching factor; and at least one channel characteristic.
Need to check novelty before this filing date? Find Prior Art