UE Mobility Preference Prediction for 5G Cell and Bandwidth Adaptation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing wireless communication systems, particularly in LTE and 5G NR, face challenges in efficiently managing UE mobility events such as secondary cell group and primary/secondary cell additions, bandwidth adaptation, and paging, which can lead to suboptimal network performance and user experience.
Innovation Solution
Implementing machine learning-based predictive models in UEs to determine preferences for network-controlled actions, carrier aggregation, dual connectivity, bandwidth part configuration, and paging strategies, allowing UEs to transmit indications to the network for tailored configurations based on past configurations and usage patterns.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional network-controlled methods are used for mobility events, then network control is maintained, but network performance and user experience are suboptimal
Solution Approach 1:
The UE autonomously determines its mobility preferences using machine learning models trained on historical configuration and usage data. The UE self-evaluates its mobility patterns, cell addition preferences, bandwidth adaptation needs, and paging strategies without requiring continuous network control, thereby improving both network performance and user experience through intelligent self-configuration.
2Productivity
If machine learning-based predictive models are implemented in UEs, then proactive and personalized network adjustments are achieved, but device complexity increases
Solution Approach 1:
The machine learning models are trained in advance using historical UE configuration data and usage patterns. This preliminary training enables the UE to make accurate predictions about its mobility preferences, cell addition needs, bandwidth requirements, and paging strategies before actual mobility events occur, improving network efficiency while keeping real-time processing complexity manageable.
Solution Approach 2:
The patent introduces a preference indication mechanism as an intermediary between the UE and network. The UE processes complex ML predictions internally and communicates simplified preference indications to the network, which then provides tailored configurations. This intermediary approach allows sophisticated ML functionality without overwhelming the communication interface or requiring complex real-time network-UE coordination.
3Adaptability or versatility
If UEs transmit preference indications to the network, then tailored configurations are received, but signaling overhead increases
Solution Approach 1:
The patent extracts only the essential preference indications from the full set of possible UE configurations and transmits these condensed signals to the network. By taking out only the critical preference information needed for tailored configuration (such as mobility pattern preferences, cell addition preferences, bandwidth adaptation preferences, and paging strategy preferences) rather than transmitting complete configuration details, the system achieves high configuration adaptability while minimizing signaling overhead.
Data Source
AI summary
Apparatuses, systems, and methods for UE mobility enhancements, including apparatuses, systems, and methods for prediction of mobility events such as secondary cell group and/or PSCell additions as well as bandwidth adaptation and paging, e.g., in 5G NR systems and beyond. A UE may determine a UE preference for a network-controlled action based on a predictive model. The predictive model may be a machine learning-based predictive model and trained based, at least in part, on past UE configurations and on past UE usage of the past UE configurations. The UE may be configured to transmit, to a network, an indication of the UE preference, e.g., via a radio resource control (RRC) message, and receive, from the network, a configuration associated with the network-controlled action that may be based, at least in part, on the indication of the UE preference.