Data Modeling with Large Random Matrices in a Cognitive Radio Network Testbed: Initial Experimental Demonstrations with 70 Nodes

Abstract

This short paper reports some initial experimental demonstrations of the theoretical framework: the massive amount of data in the large-scale cognitive radio network can be naturally modeled as (large) random matrices. In particular, using experimental data we will demonstrate that the empirical spectral distribution of the large sample covariance matrix---a Hermitian random matrix---agree with its theoretical distribution (Marchenko-Pastur law). On the other hand, the eigenvalues of the large data matrix ---a non-Hermitian random matrix---are experimentally found to follow the single ring law, a theoretical result that has been discovered relatively recently. To our best knowledge, our paper is the first such attempt, in the context of large-scale wireless network, to compare theoretical predictions with experimental findings.

4 pages, 11 figures

References in corpus (2)

• Outliers in the Single Ring Theorem
• Universality of sample covariance matrices: CLT of the smoothed empirical spectral distribution

Cited by in corpus (5)

• A Data Driven Approach for Resting-state EEG signal Classification of Schizophrenia with Control Participants using Random Matrix Theory
• Outliers in the Single Ring Theorem
• The Foundation of Big Data: Experiments, Formulation, and Applications
• Massive MIMO as a Big Data System: Random Matrix Models and Testbed
• A Novel Approach for Big Data Analytics in Future Grids Based on Free Probability