5G Link Selection Using ML for Uplink and Capacity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In non-standalone 5G networks, existing link selection methods are limited by focusing only on downlink coverage, ignoring uplink limitations, cell capacity, and dynamic link fluctuations, leading to inefficient resource allocation and increased signaling costs.
Innovation Solution
A method using a machine learning model to predict the optimal NR station for user equipment by considering channel quality measurements such as cell range, uplink performance, cell capacity, and link quality, as well as historical and real-time data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If traditional link selection methods are used focusing only on downlink coverage, then downlink coverage is improved, but uplink performance and resource allocation efficiency deteriorate
Solution Approach 1:
The patent changes the selection parameters from only downlink coverage metrics to multiple parameters including uplink performance, cell capacity, and link quality measurements, thereby resolving the contradiction between downlink coverage and uplink performance
Solution Approach 2:
The patent adds new dimensions to the link selection process by incorporating cell capacity, uplink performance, and dynamic link quality measurements beyond the traditional single-dimension downlink coverage approach, enabling comprehensive optimization
2Device complexity
If static link selection is used, then device complexity is reduced, but resource allocation efficiency and adaptability to dynamic conditions deteriorate
Solution Approach 1:
The patent implements dynamic link selection by continuously monitoring link quality measurements and cell capacity, allowing the system to adapt to changing network conditions in real-time, thereby resolving the contradiction between simplicity and efficiency
Solution Approach 2:
The patent introduces feedback mechanisms where link quality measurements and performance data are continuously collected and used to adjust link selection decisions, enabling efficient resource allocation while maintaining manageable system complexity through automated closed-loop control
3Measurement precision
If comprehensive channel quality measurements are collected, then link selection accuracy is improved, but signaling costs and network overhead increase
Solution Approach 1:
The patent applies partial measurement by selecting specific key parameters (uplink performance, cell capacity, link quality) rather than measuring all possible channel characteristics, thereby achieving sufficient link selection accuracy while minimizing signaling overhead
Data Source
AI summary
A method, a system, and a computer program product for optimal link selection between user equipment and new radio stations in a non-standalone network. The method includes detecting a connection attempt by a user equipment to a non-standalone network that includes at least one base station and NR stations. The method also includes receiving connection data from the UE and connecting the UE to the base station. The method further includes determining an NR station from the NR stations based on channel quality measurements relating to the NR stations and the connection data of the UE using a prediction module that uses a machine learning model to predict the NR station. The channel quality measurements include cell range, uplink performance, cell capacity, and link quality. The method also includes connecting the UE to the NR station determined by the machine learning model.