Mutual Information-Driven Visualization and Clustering for Core KPI Selection in O-RAN Testing

2025-10-02

AI Summary

To address the exponential growth of performance metrics in O-RAN systems—which drastically increases testing complexity—this paper proposes an information-theoretic, automated KPI selection method. The approach models multidimensional time series as stochastic processes and introduces frequency-domain Aggregated Mutual Information (AMIF) as a robust surrogate for directed information; it further employs quantile-based estimation to enhance small-sample robustness. Multidimensional scaling (MDS) is integrated for visualizing dependency structures, while DBSCAN enables density-based clustering to automatically identify strongly correlated KPI clusters. This work represents the first systematic application of an information-theoretic framework to core KPI discovery in O-RAN. Evaluated on real-world test data, the method successfully extracts representative KPI sets—such as those governing link adaptation—reducing both dimensionality and computational overhead for subsequent learning-driven testing and optimization.

Technology Category

• Data Mining & Knowledge Management: Anomaly/Outlier Detection
• Search and Optimization: Algorithm Configuration
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Web performance, measurement, and characterization
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Security and Privacy: Large-scale security measurements

Abstract

O-RAN testing is becoming increasingly difficult with the exponentially growing number of performance measurements as the system grows more complex, with additional units, interfaces, applications, and possible implementations and configurations. To simplify the testing procedure and improve system design for O-RAN systems, it is important to identify the dependencies among various performance measurements, which are inherently time-series and can be modeled as realizations of random processes. While information theory can be utilized as a principled foundation for mapping these dependencies, the robust estimation of such measures for random processes from real-world data remains challenging. This paper introduces AMIF-MDS, which employs aggregate mutual Information in frequency (AMIF), a practical proxy for directed information (DI), to quantify similarity and visualize inter-series dependencies with multidimensional scaling (MDS). The proposed quantile-based AMIF estimator is applied to O-RAN time-series testing data to identify dependencies among various performance measures so that we can focus on a set of ``core'' performance measures. Applying density-based spatial clustering of applications with noise (DBSCAN) to the MDS embedding groups mutually informative metrics, organically reveals the link-adaptation indicators among other clusters, and yields a ``core'' performance measure set for future learning-driven O-RAN testing.

Problem

• Identify dependencies among O-RAN performance measurements using mutual information
• Simplify testing by selecting core KPIs through clustering techniques
• Visualize time-series dependencies with multidimensional scaling for system optimization

Innovation

• Uses AMIF as proxy for directed information
• Applies multidimensional scaling for dependency visualization
• Employs DBSCAN clustering to identify core metrics