An approach to eliminating end effects of EMD through mirror extension coupled with support vector machine method

摘要

The mirror extension is a basic algorithm to treat the end effects in the empirical mode decomposition (EMD) of signals. It must meet the requirements of the specular position at the local extremum, but the actual signal is very difficult to implement. For this reason, its decomposition can lead to severe distortion. This paper proposed a new approach to the performance improvement of end effect elimination in EMD method through the data extension on the basis of traditional mirror extension technique coupled with the function regression method of support vector machine (SVM). Some data outside of both ends of an original signal are firstly predicted by means of the relationships obtained by the function regression method of SVM, from which one or more extreme points outside each end are captured. And then the mirror extension algorithm is used to inhibit the end effects possibly occurring in operation of EMD method. The application examples of the simulated signal show that the proposed method can effectively eliminate the end effect of the EMD method.