A Hypothesis Testing Approach to Nonstationary Source Separation

TL;DR

This paper introduces a hypothesis testing framework for nonstationary source separation, providing a general approach that improves the identification and extraction of nonstationary signals, demonstrated through fetal ECG case study.

Contribution

A new hypothesis testing-based framework for nonstationary source separation that offers a unified classification approach, surpassing problem-specific methods.

Findings

Effective separation of nonstationary signals demonstrated on fetal ECG data

Framework generalizes existing methods for nonstationarity detection

Improves robustness and applicability of source separation techniques

Abstract

The extraction of nonstationary signals from blind and semi-blind multivariate observations is a recurrent problem. Numerous algorithms have been developed for this problem, which are based on the exact or approximate joint diagonalization of second or higher order cumulant matrices/tensors of multichannel data. While a great body of research has been dedicated to joint diagonalization algorithms, the selection of the diagonalized matrix/tensor set remains highly problem-specific. Herein, various methods for nonstationarity identification are reviewed and a new general framework based on hypothesis testing is proposed, which results in a classification/clustering perspective to semi-blind source separation of nonstationary components. The proposed method is applied to noninvasive fetal ECG extraction, as case study.