TY - JOUR

T1 - Usefulness of Quasi-Ll norm-based non-negative matrix factorization algorithm to estimate background signal using environmental electromagnetic field measurements at ELF band

AU - Mouri, Motoaki

AU - Takumi, Ichi

AU - Yasukawa, Hiroshi

AU - Cichocki, Andrzej

PY - 2016

Y1 - 2016

N2 - Our research group has been measuring Extremely Low Frequency (ELF) magnetic fields across Japan. The ELF measurements are mixtures of signals associated with various natural or artificial phenomena. When focus on specific factor, the signals related to other factors distort analysis result. In order to get specific information accurately, we should estimate desired signals or eliminate undesired signals. We found that Image Space Reconstruction Algorithm (ISRA), one of the Non-negative Matrix Factorization (NMF) algorithm, works better than independent component analysis to estimate the ELF background signal. However, ISRA sometimes failed to estimate the weight vector for the background signal. We considered that ISRA has weakness for outliers and sparse signals because ISRA is based on minimizing L2 (Frobenius) norm between input matrix and projected matrix from estimated matrices. In order to improve robustness, we developed new methods based on minimizing quasi-Ll norm (QL1-NMF). In the experiment using generated signals and ELF observed signals which had outliers, the proposed method estimate background signal more accurately than ISRA and other LI norm based algorithms.

AB - Our research group has been measuring Extremely Low Frequency (ELF) magnetic fields across Japan. The ELF measurements are mixtures of signals associated with various natural or artificial phenomena. When focus on specific factor, the signals related to other factors distort analysis result. In order to get specific information accurately, we should estimate desired signals or eliminate undesired signals. We found that Image Space Reconstruction Algorithm (ISRA), one of the Non-negative Matrix Factorization (NMF) algorithm, works better than independent component analysis to estimate the ELF background signal. However, ISRA sometimes failed to estimate the weight vector for the background signal. We considered that ISRA has weakness for outliers and sparse signals because ISRA is based on minimizing L2 (Frobenius) norm between input matrix and projected matrix from estimated matrices. In order to improve robustness, we developed new methods based on minimizing quasi-Ll norm (QL1-NMF). In the experiment using generated signals and ELF observed signals which had outliers, the proposed method estimate background signal more accurately than ISRA and other LI norm based algorithms.

KW - Environmental electromagnetic fields

KW - LI norm

KW - Non-negative matrix factorization

KW - Signal separation

UR - http://www.scopus.com/inward/record.url?scp=84971229231&partnerID=8YFLogxK

U2 - 10.1541/ieejfms.136.241

DO - 10.1541/ieejfms.136.241

M3 - Article

AN - SCOPUS:84971229231

VL - 136

SP - 241

EP - 251

JO - IEEJ Transactions on Fundamentals and Materials

JF - IEEJ Transactions on Fundamentals and Materials

SN - 0385-4205

IS - 5

ER -