ML Component Coordination in Wireless UEs and Network Nodes

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

In wireless communication systems, network nodes often signal changes in settings without awareness of appropriate configurations for user equipment, leading to inefficient or inaccurate machine-learning-based operations that negatively impact network performance due to uncorrelated machine learning components.

Innovation Solution

Implement techniques for managing changes to machine learning models and parameters by transmitting UE capability information and change indications between network nodes and user equipment, ensuring correlated outputs through UE and network node collaboration.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If network nodes signal changes in settings without awareness of appropriate configurations, then network node autonomy is maintained, but machine learning component correlation deteriorates leading to inefficient operations

Engineering Contradiction:

Improvenetwork node autonomyVSAvoidmachine learning component correlation

Core Design Contradiction:

Adaptability or versatilityVSReliability

Solution Approach 1:

The patent implements feedback mechanisms where UEs report their machine learning component configurations and capability information back to the network node. This allows the network node to adjust its settings based on actual UE states, maintaining correlation between network and UE machine learning components while preserving network node autonomy through informed decision-making.

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The patent employs preliminary action by having UEs transmit capability information indicating their machine learning component states before the network node signals configuration changes. This advance knowledge allows the network node to plan configuration changes that maintain correlation, preventing inefficient operations before they occur.

Inventive Principle:

Principle #10Preliminary action

2Adaptability or versatility

If machine learning component configurations are changed without coordination, then configuration flexibility increases, but operational accuracy deteriorates

Engineering Contradiction:

Improveconfiguration flexibilityVSAvoidoperational accuracy

Core Design Contradiction:

Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The system uses feedback loops where UEs report their capability information and the network node acknowledges configuration changes. This coordinated approach allows flexible configuration changes while maintaining operational accuracy through continuous verification of correlation between network and UE machine learning components.

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The patent implements preliminary action by requiring UEs to transmit capability information before configuration changes are applied. This preliminary exchange of information ensures that both network node and UE have synchronized understanding of the configuration state, maintaining accuracy while enabling flexibility.

Inventive Principle:

Principle #10Preliminary action

3Reliability

If change notification mechanisms are implemented, then machine learning component correlation improves, but signaling overhead increases

Engineering Contradiction:

Improvemachine learning component correlationVSAvoidsignaling overhead

Core Design Contradiction:

ReliabilityVSLoss of information

Solution Approach 1:

The patent extracts only the essential capability information and configuration state data that are necessary for maintaining machine learning component correlation. By transmitting only this relevant subset of information rather than complete configuration details, the system maintains correlation while minimizing signaling overhead.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The system applies partial action by implementing change notifications only when actual configuration changes occur that affect machine learning component correlation. This selective notification approach maintains reliability by notifying of relevant changes while avoiding excessive signaling for irrelevant updates.

Inventive Principle:

Principle #16Partial or excessive action

Data Source

PatentUS12556975B2Machine learning component management

Publication Date: 2026.02.17 QUALCOMM INC

AI summary

Various aspects of the present disclosure generally relate to wireless communication. In some aspects, a user equipment (UE) may transmit, to a network node, UE capability information associated with at least one machine learning component. The UE may receive, from the network node and based on the UE capability information, configuration information corresponding to the at least one machine learning component. The UE may generate a first machine learning output based on the machine learning component. The UE may perform a communication task based on the first machine learning output. Numerous other aspects are described.