Removal of noise from the body-seat interface temperature signal using empirical mode decomposition

Liu, Z., Zhongming, L., Heusch, A.I., Cascioli, V. and McCarthy, P.W. (2010) Removal of noise from the body-seat interface temperature signal using empirical mode decomposition. In: 3rd International Conference on Computational Intelligence and Industrial Application (PACIIA 2010), 4-5 December 2010, Wuhan, China.

Abstract

This paper applied the data-driven technique of Empirical Mode Decomposition (EMD) to smooth raw thermal data between body and seat interface. The performance of this EMD-based filter was compared with Moving Av erage (MA) filter, Local Regression (LR) filter and Savitzky-Golay (SG) filter using the goodness of statistics and error charts as judgment criteria. Results showed the EMD-based filter outperformed other de-noising algorithms with the lowest RMSE (root-mean-square-error) and the highest R 2 values in the application of curve fitting based on the two power series.