UE AI Handover Prediction for Secondary Node Change Reliability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing wireless communication systems, particularly in 5G NR, face challenges in predicting handover and secondary node change failures, leading to suboptimal network operations due to incomplete measurement reports from wireless devices.
Innovation Solution
Implementing artificial intelligence/machine learning (AI/ML) at the UE level to predict handover and PSCell update success probabilities by measuring candidate cells, SSBs, and RSs, and providing these predictions to the network for informed decision-making.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional measurement reports are used for handover decisions, then network operations can be maintained with existing infrastructure, but handover failure prediction accuracy is insufficient
Solution Approach 1:
The UE performs self-service by autonomously executing AI/ML models to predict handover outcomes using its own measurement data, eliminating the need for complex network-side processing and enabling accurate predictions directly at the UE without requiring additional network infrastructure
Solution Approach 2:
The patent replaces traditional mechanical measurement reporting mechanisms with AI/ML-based prediction models that process measurement data to forecast handover failures, substituting simple data collection with intelligent analysis capabilities
2Measurement precision
If more measurement data is collected from candidate cells, then prediction accuracy improves, but measurement time and processing overhead increase
Solution Approach 1:
The UE performs preliminary actions by pre-configuring AI/ML models and preparing measurement data structures before handover decisions are needed, enabling rapid predictions without requiring extensive real-time data collection during critical decision moments
Solution Approach 2:
The patent applies parameter changes by dynamically adjusting the scope and depth of measurements based on current network conditions and prediction requirements, collecting only the necessary measurement parameters for accurate predictions rather than all possible data
3Reliability
If handover predictions are made more accurately, then network reliability improves, but the complexity of prediction models increases
Solution Approach 1:
The UE performs self-service by autonomously executing AI/ML models to predict handover outcomes using its own measurement data, eliminating the need for complex network-side processing and enabling accurate predictions directly at the UE without requiring additional network infrastructure
Solution Approach 2:
The patent applies parameter changes by dynamically adjusting the scope and depth of measurements based on current network conditions and prediction requirements, collecting only the necessary measurement parameters for accurate predictions rather than all possible data
Data Source
AI summary
The apparatus may be a wireless device configured to obtain information related to at least one of a plurality of candidate cells and a source cell, a plurality of synchronization signal blocks (SSBs), or a plurality of reference signals (RSs) to measure in association with a UE mobility, perform, based on the information, a set of measurements of at least one of the plurality of candidate cells and the source cell, the plurality of SSBs, or the plurality of RSs, transmit, for a first network device, a set of predictions related to at least one potential handover operation with at least one candidate cell in the plurality of candidate cells based on the set of measurements performed by the UE and associated with the UE mobility, and perform a handover operation based on the set of predictions.