Anzeige
Erschienen in:
2019 | OriginalPaper | Buchkapitel
Efficiency of AR, MA and ARMA Models in Prediction of Raw and Filtered Center of Pressure Signals
verfasst von : Guilherme Augusto Gomes De Villa, Anderson Juliano Silvestre, Gustavo Souto de Sá e Souza, Adriano Carafini, Alfredo de Oliveira Assis, Adriano O. Andrade, Marcus Fraga Vieira
Erschienen in: XXVI Brazilian Congress on Biomedical Engineering
Verlag: Springer Singapore
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
Abstract
In this study the efficiency of the AR, MA and ARMA models was analyzed for the prediction of a center of pressure signal in a quiet upright stance on a force platform. The analysis was performed to verify differences among the above models to predict the next sample of a signal. A small prediction error allow to predict and replace any gaps in the acquired signals, which is interesting in any area of science, especially when it comes from biological signals in its most varied aspects. The Welch method was used for spectral estimation. The prediction error presented the greatest variation when comparing the cases with raw (unfiltered) and filtered data. We observed that using these models on unfiltered signals resulted in poor prediction; however, the results indicated ARMA modeling as a good predictor using filtered center of pressure signals.