Pavement profile analysis using ensemble empirical mode decomposition

Abstract

Pavement profile analysis is a major component in pavement infrastructure management decision making for maintenance and rehabilitation. Road profile data are non-stationary and inherently non-Gaussian. This paper presents the application ensemble empirical mode decomposition (EEMD) to profile data analysis. The EEMD employs multiple applications of empirical mode decomposition (EMD), to which white noise is added, with constant amplitude relative to the target data. The uniform distribution of white noise over the entire time frequency space provides a reference frame for signals of comparative scale to collate in one intrinsic mode function, thus eliminating mode mixing. The EEMD approach was compared with the traditional EMD and it appears that the EEMD outperforms the traditional EMD.