Robust Multi-dimensional Model Order Estimation Using LineAr Regression of Global Eigenvalues (LaRGE)

TL;DR

The paper introduces LaRGE, a robust method for estimating the model order of noisy multi-dimensional low-rank data using global eigenvalues derived from HOSVD, outperforming existing techniques especially in biomedical applications.

Contribution

LaRGE is a novel model order estimation technique that does not require false alarm probability calculations and offers improved accuracy over current methods.

Findings

LaRGE outperforms state-of-the-art methods in simulations.

LaRGE effectively analyzes EEG data.

The method is robust to noise in multi-dimensional data.

Abstract

The efficient estimation of an approximate model order is very important for real applications with multi-dimensional data if the observed low-rank data is corrupted by additive noise. In this paper, we present a novel robust method for model order estimation of noise-corrupted multi-dimensional low-rank data based on the LineAr Regression of Global Eigenvalues (LaRGE). The LaRGE method uses the multi-linear singular values obtained from the HOSVD of the measurement tensor to construct global eigenvalues. In contrast to the Modified Exponential Test (EFT) that also exploits the approximate exponential profile of the noise eigenvalues, LaRGE does not require the calculation of the probability of false alarm. Moreover, LaRGE achieves a significantly improved performance in comparison with popular state-of-the-art methods. It is well suited for the analysis of biomedical data. The excellent performance of the LaRGE method is illustrated via simulations and results obtained from EEG recordings.