RIC KPI Prediction Using GNN Surrogates for Faster Network Testing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing RIC simulation processes for network testing are time-consuming and resource-intensive, requiring extensive repetitions for different configuration files, which hinders efficient testing and optimization of RIC functionalities.
Innovation Solution
A network test device employs an AI/ML model, specifically a GNN, trained on a labeled dataset generated through parallel RIC simulations of distinct configuration files, to predict key performance indicators (KPIs) such as energy savings potential, reducing the need for extensive simulations and resource consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional RIC simulation processes are used for network testing, then measurement precision and reliability are improved, but loss of time and productivity deteriorate due to extensive repetitions required
Solution Approach 1:
The system performs preliminary RIC simulations to generate a labeled dataset before actual testing, training an AI/ML model in advance. This preliminary action creates a predictive model that can quickly estimate KPIs without requiring full simulation repetitions during testing, thus reducing time loss while maintaining accuracy through the pre-trained model's predictions
Solution Approach 2:
The system creates a copy of the RIC simulation process by training an AI/ML model on simulation data. This digital twin or surrogate model replicates the behavior of the actual RIC system, allowing rapid predictions of KPIs without executing time-consuming physical simulations, thereby resolving the contradiction between prediction accuracy and time consumption
2Measurement precision
If extensive RIC simulations are performed for different configuration files, then measurement precision is improved, but productivity and use of energy deteriorate due to resource-intensive repetitions
Solution Approach 1:
The system performs preliminary simulations to build a trained AI/ML model that can predict energy savings and KPIs rapidly. This preliminary action eliminates the need for repeated extensive simulations during actual testing, significantly improving productivity while maintaining measurement precision through the pre-trained model's accurate predictions
Solution Approach 2:
The system creates a surrogate model that copies the essential predictive capabilities of the full RIC simulation system. This copied model enables rapid evaluation of different configuration files and xApps without requiring resource-intensive repeated simulations, thus resolving the contradiction between prediction accuracy and testing efficiency
3Reliability
If multiple repetitions of RIC simulations are performed, then reliability of KPI measurement is improved, but loss of time and productivity worsen
Solution Approach 1:
The system performs preliminary simulations to train an AI/ML model that captures the reliable relationships between network configurations and KPIs. This preliminary action establishes a predictive model that can quickly provide reliable KPI estimates without requiring multiple repeated simulations, thus improving productivity while maintaining measurement reliability through the trained model's consistent predictions
Data Source
AI summary
In some implementations, a network test device may obtain a labeled dataset based on a radio access network intelligent controller (RIC) test simulation. The network test device may generate an artificial intelligence or machine learning (AI/ML) model that is trained using the labeled dataset. The network test device may receive, via a user interface, a network configuration associated with a user application deployed on an RIC. The network test device may predict, using the AI/ML model running on the network test device, a key performance indicator (KPI) based on the network configuration. The network test device may provide, via the user interface, an indication of the KPI.