AI/ML Model Identification via Reference Signals for UE Configuration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing AI/ML model switching in wireless communications systems requires substantial coordination between network and user equipment vendors, lacking a clear framework for functionality-based model switching and signaling, especially in environments with varying configurations.
Innovation Solution
Implement a functionality-based AI/ML model switching framework using a reference signal transmission and measurement phase to facilitate model switching based on model functionality, independent of public model IDs, enabling robust coordination across different network and UE vendors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If AI/ML model switching is implemented with substantial coordination between network and UE vendors, then model switching reliability is improved, but system complexity and deployment difficulty increase
Solution Approach 1:
The patent introduces a functionality-based identification mechanism as an intermediary layer between the network and UE. Instead of directly coordinating on model structures, the system uses reference signals with associated functionality identifiers to mediate model switching. The UE measures reference signals, identifies the corresponding functionality, and selects the appropriate AI/ML model automatically, eliminating the need for substantial vendor coordination while maintaining reliability.
Solution Approach 2:
The UE performs self-service by autonomously measuring reference signals, identifying the functionality based on the measurements, and selecting the appropriate AI/ML model without requiring network-side coordination on model-specific parameters. The UE independently completes the model selection process based on the functionality identification, reducing the coordination burden between network and UE vendors.
2Adaptability or versatility
If functionality-based model switching is implemented without public model IDs, then adaptability across different vendors is improved, but model identification precision may be compromised
Solution Approach 1:
The patent changes the identification parameter from public model IDs to functionality-based identifiers associated with reference signals. Instead of identifying models by their public IDs, the system identifies models through the functionality parameters of the reference signals (such as beam direction, signal type, and transmission characteristics). This parameter change enables vendor-agnostic model identification while maintaining precision through the measurement of reference signal properties.
Solution Approach 2:
The patent creates a functional copy of the model identification process. Rather than using direct model IDs, the system measures the reference signal properties and creates a functional representation of the model's intended behavior. This functional copy allows the UE to identify and select the correct model based on its functionality rather than its identifier, achieving both adaptability and precision.
3Reliability
If AI/ML models are switched based on channel properties and reference signal measurements, then communication performance is improved, but measurement and processing time increase
Solution Approach 1:
The patent performs preliminary action by pre-configuring reference signals with associated functionality identifiers and pre-establishing the mapping between reference signal properties and AI/ML model characteristics. The UE measures the reference signals and directly identifies the functionality without needing to perform complex real-time analysis. This preliminary setup reduces the measurement and processing time required for model switching while maintaining communication performance.
Data Source
AI summary
Various aspects of the present disclosure relate to AI/ML model identifier acquisition for AI/ML-based model configuration. An apparatus, such as a UE, receives, from a network entity, an artificial intelligence (AI)-based configuration corresponding to signal transmission and/or signal reception by the UE, where the AI-based configuration is associated with one or more AI models, and an AI model is associated with a dataset. The UE receives, from the network entity, a reference signal configuration for a set of reference signals, and the reference signal configuration is paired with the AI-based configuration. The UE measures a set of report parameters based at least in part on the set of reference signals. The UE transmits, to the network entity, a feedback report comprising the set of report parameters, and the feedback report is usable by the network entity to select the AI model from the one or more AI models.