We propose an extension for Empirical Mode Decomposition (EMD) that works for multivariate functions defined on scattered finite dimensional domains and extend this to the case of vector-valued functions using principal curves. After the decomposition is performed, we classify the obtained modes using a vector that measures the average period of oscillation in each direction. Finally, we show an interesting application to a spatio-temporal problem on population dynamics for Dengue Haemorrhagic Fever on Tailand
Empirical mode decomposition on scattered datasets
VENTURINO, Ezio
2008-01-01
Abstract
We propose an extension for Empirical Mode Decomposition (EMD) that works for multivariate functions defined on scattered finite dimensional domains and extend this to the case of vector-valued functions using principal curves. After the decomposition is performed, we classify the obtained modes using a vector that measures the average period of oscillation in each direction. Finally, we show an interesting application to a spatio-temporal problem on population dynamics for Dengue Haemorrhagic Fever on TailandFile in questo prodotto:
Non ci sono file associati a questo prodotto.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.