Data driven time-frequency analysis based on empirical mode decomposition and adaptive optimal kernel

摘要

Hilbert-Huang transfer (HHT) and adaptive optimal kernel (AOK) are data driven time frequency analysis algorithm. But HHT is limited by Bedrosian theorem and AOK normally behaviors well only for single component signal. To resolve above problems, empirical mode decomposition (EMD), kernel part of HHT, and AOK are combined together to create a new time-frequency representation (TFR). So by this novel TFR more extensive type signal can be analyzed, which are difficult to be processed by HHT or AOK individually in the past. EMD is used to decompose multicomponent signal into a bundle of single component signals and then AOK is applied to compute the TFR of individual single component, finally all these TFRs are summed together to generate one TFR. The new TFR shares the unique features from EMD and AOK, and minimizes their defects. The results of examples verify that TFR based on EMD and AOK is practical.