Flexible MDT Configuration for AI/ML Data Collection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current Minimization of Driving Test (MDT) configurations are inflexible, limiting network nodes' ability to adapt measurement configurations for AI/ML model training and performance monitoring, which requires different data collection settings than those provided by the network entity.
Innovation Solution
An apparatus and method that allow a network node to receive an original MDT configuration and an indication of allowed modifications, enabling the node to configure a modified MDT configuration for user devices, which can include changes to logging intervals, measurement sets, and reporting triggers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a fixed original MDT configuration is provided by the network entity, then the configuration structure is simple and easy to manage, but the network node cannot adapt the measurement configuration for AI/ML model training and performance monitoring
Solution Approach 1:
The patent segments the MDT configuration into a fixed original configuration template and variable modification parameters. The network entity provides the original configuration structure, while the network node can selectively modify specific parameters (logging intervals, measurement sets, reporting triggers) based on AI/ML training needs, achieving adaptability without complete redesign of the configuration system
Solution Approach 2:
The patent introduces dynamic modification capability to the previously static MDT configuration. The network node receives the original configuration and can dynamically adjust parameters such as logging intervals, measurement sets, and reporting triggers according to real-time AI/ML model training requirements, transforming the configuration from fixed to adaptable
2Measurement precision
If the network node modifies the MDT configuration to collect data tailored for AI/ML training, then the data collection accuracy improves, but the configuration management becomes more complex
Solution Approach 1:
The patent applies local quality modification by allowing the network node to adjust only specific parameters of the MDT configuration (such as logging intervals, measurement sets, reporting triggers) while maintaining the overall configuration structure. This enables tailored data collection for AI/ML training without requiring complete redesign of the configuration management system
Solution Approach 2:
The patent implements a feedback mechanism where the network node receives the original MDT configuration, modifies it according to AI/ML training needs, and sends the modified configuration back to the network entity. This feedback loop allows for accurate data collection while maintaining configuration management through standardized communication protocols
Data Source
AI summary
Embodiments of the present disclosure relate to devices, methods, apparatuses and computer readable storage media related to minimization of driving test configuration. The method comprises receiving an original MDT configuration, receiving an indication indicating whether all or a part of the original MDT configuration is allowed to be modified, configuring a modified MDT configuration from the original MDT configuration based on the indication, and transmitting the modified MDT configuration.