Multi-TRP Associated ID Mapping for AI/ML Applicability

The patent segments the identification process into multiple hierarchical levels using associated IDs and sub-associated IDs. Each level represents different granularities of TRP configurations, functionalities, and transmission scenarios. This segmentation enables precise identification and data collection for AI/ML model applicability determination, resolving the information loss problem while maintaining improved network performance.

The patent introduces associated IDs and sub-associated IDs as intermediary elements that bridge the gap between network conditions and AI/ML model applicability. These intermediaries facilitate systematic data collection and identification, enabling the network to efficiently determine model applicability without losing critical information, thus resolving the contradiction between performance improvement and information loss.

The patent implements a dynamic applicability determination mechanism where the WTRU evaluates associated IDs against multiple sub-associated IDs based on current network conditions. The system dynamically determines whether an associated ID is fully applicable, partially applicable, or not applicable by checking against configurable thresholds. This dynamic approach maintains versatility while managing complexity through adaptive evaluation rather than static rules.

The WTRU autonomously performs the applicability determination process by evaluating configuration information and comparing it against stored associated IDs and sub-associated IDs. The device self-services by independently determining which AI/ML models are applicable without requiring complex external coordination, thus improving versatility while keeping the determination process manageable through localized intelligence.