Applying Empirical Mode Decomposition and mutual information to separate stochastic and deterministic influences embedded in signals

Author(s)

Rios, Ricardo Araujo ^[1] ; de Mello, Rodrigo Fernandes ^[2]

Affiliation

^[1] Univ Fed Bahia, Dept Comp Sci, BR-41170290 Salvador, BA - Brazil

^[2] Univ Sao Paulo, Inst Math & Comp Sci, BR-05508 Sao Paulo - Brazil

• Document type: Journal article
• Source: Signal Processing; v. 118, p. 159-176, JAN 2016.
• Web of Science Citations: 12

Abstract

Empirical Mode Decomposition (EMD) is a method to decompose signals into Intrinsic Mode Functions (IMB) to be analyzed in terms of instantaneous frequencies and amplitudes. By comparing the phase spectra of IMEs, we observed that a subset of them contains more stochastic influences while the other is predominantly deterministic. Considering this observation, we claim that IMEs can be combined to form two additive components: one deterministic and another stochastic. Having both components separated, researchers can improve data modeling as well as forecasting. In this context, this paper presents a new approach to separate deterministic from stochastic influences embedded in signals, considering the mutual information contained in phase spectra of consecutive IMFs. As previous step of this study, we also proved that EMD works as a filter bank. (C) 2015 Elsevier B.V. All rights reserved. (AU)

• FAPESP's process: 14/13323-5 - An approach based on the stability of clustering algorithms to ensure concept drift detection on data streams
• Grantee: Rodrigo Fernandes de Mello
• Support Opportunities: Regular Research Grants
• FAPESP's process: 09/18293-9 - A Hybrid Approach to Identify and Model Deterministic and Stochastic Components present in Time Series
• Grantee: Ricardo Araújo Rios
• Support Opportunities: Scholarships in Brazil - Doctorate