WTRU Predicted Measurement Reporting for Proactive Event Triggers

The patent applies preliminary action by predicting measurement events in advance using AI/ML models and reporting them before they actually occur. The WTRU determines predicted measurements at future time points and sends reports indicating these predicted events, allowing the network to prepare resources proactively rather than reacting to measurements after they occur.

The system implements feedback mechanisms where the WTRU receives configuration information including time offsets and trigger conditions from the network, processes these configurations through AI/ML models, and adjusts reporting behavior accordingly. The network receives predicted measurement reports and can modify future configurations based on the accuracy and utility of the predictions.

Instead of continuously reporting measurements in real-time, the system performs preliminary prediction of when measurement events are likely to occur and reports only at those predicted moments. This reduces the frequency of reports while maintaining timely information about network conditions, thereby reducing energy consumption while avoiding significant time loss.

The patent changes the parameter of measurement reporting from continuous real-time reporting to event-driven prediction-based reporting. By using AI/ML models to predict when threshold conditions will be met, the system adjusts the reporting timing parameter to send messages only when necessary, optimizing the balance between response time and energy consumption.

The system uses feedback from network configurations and predicted measurement results to refine future predictions. The WTRU receives configuration information including trigger conditions and time offsets, processes these through AI/ML models, and adjusts prediction thresholds dynamically. This feedback loop allows the system to achieve high measurement precision without requiring manually configured complex thresholds.

The WTRU performs self-service by automatically determining prediction thresholds and confidence levels using embedded AI/ML models. Rather than requiring network-side configuration of complex threshold parameters, the WTRU autonomously processes configuration information and adapts its reporting behavior, reducing the complexity burden on network operators while maintaining high prediction accuracy.