Radio Access Network Model Evaluation with AMF-Coordinated Measurements
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies face challenges in efficiently evaluating and optimizing the performance of radio access networks, particularly in 5G systems, due to the complexity and dynamic nature of wireless communication environments, leading to suboptimal resource allocation and user experience.
Innovation Solution
Implementing artificial intelligence and machine learning models to analyze network performance data and optimize resource allocation in real-time, adapting to traffic patterns and device capabilities, thereby enhancing network efficiency and user experience.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional network optimization methods are used, then implementation simplicity is maintained, but network performance optimization is insufficient due to environment complexity and dynamics
Solution Approach 1:
The patent introduces an intermediary evaluation system comprising a data collection module, feature extraction module, and model evaluation module that mediates between the complex wireless environment and traditional optimization methods. This intermediary layer processes raw network data through multiple modules to generate comprehensive performance evaluations, resolving the contradiction by adding structured complexity that enables better performance optimization while managing system complexity through modular design
Solution Approach 2:
The evaluation system is segmented into distinct functional modules: data collection module (220), feature extraction module (222), and model evaluation module (224). Each module handles specific tasks independently, allowing the system to manage complexity through division of labor while achieving comprehensive network performance evaluation through the coordinated operation of these segmented components
2Productivity
If AI/ML models are deployed for real-time optimization, then resource allocation efficiency is improved, but computational resource consumption increases
Solution Approach 1:
The system performs preliminary feature extraction (222) on network data before feeding it to AI/ML models for optimization decisions. By pre-processing data to extract only relevant features and reducing data dimensionality beforehand, the system enables efficient real-time model execution while minimizing computational energy consumption during critical optimization operations
Solution Approach 2:
The evaluation system dynamically adapts its operation based on network conditions and resource availability. The model evaluation module (224) selectively applies different evaluation strategies and model complexities depending on current network state, allowing the system to optimize resource allocation efficiency while dynamically adjusting computational energy consumption to match actual network needs
3Measurement precision
If comprehensive network data is collected for accurate evaluation, then measurement precision is improved, but data processing time increases
Solution Approach 1:
The feature extraction module (222) extracts only the most relevant features from comprehensive network data, separating essential performance indicators from redundant information. This extraction process maintains measurement precision by focusing on critical features while significantly reducing data processing time by eliminating unnecessary data elements before model evaluation
Solution Approach 2:
The system applies partial action by collecting comprehensive network data but processing only the essential subset through feature extraction. The model evaluation module (224) uses this extracted feature subset for evaluation, achieving accurate performance measurement without the time cost of processing the entire comprehensive dataset, thus balancing precision and processing time
Data Source
AI summary
A method can include receiving, by a base station from an access and mobility management function (AMF), one or more first messages that include a model identifier of a first model and also include a data request for a model evaluation of the first model. The method can further include sending, by the base station to the AMF, one or more second messages that include the model identifier of the first model, predicted data determined based on the first model for a wireless device, and an identifier of the evaluation data. The method can additionally include sending, by the base station to the wireless device, one or more third messages requesting a measurement. The one or more third messages may include the model identifier of the first model and the identifier of the evaluation data.