Process method for end effects of Hilbert-Huang transform based on support vector regression machines

The end effects of Hilbert-Huang transform are shown in two aspects. On one hand, the end effects are produced when the signal is decomposed by empirical mode decomposition (EMD) method; on the other hand, the end effects are produced too when the Hilbert transforms are applied to the intrinsic mode functions (IMFs). To overcome the end effects of Hilbert-Huang transform, the support vector regression machines are used to predict the signal before the signal is decomposed by EMD (Empirical Mode Decomposition), thus the end effects could be overcome effectively and the IMFs (Intrinsic Mode Functions) with physical sense could be obtained. After that, to restrain the end effects of Hilbert transform, the support vector regression machines are used again to predict the IMFs before the Hilbert transform of the IMFs, therefore, the accurate instantaneous frequencies and instantaneous amplitudes could be obtained and the Hilbert spectrum with physical sense could be acquired. The analysis results from the simulated and practical signals demonstrate that the end effects of Hilbert-Huang transform could be resolved effectively by the time series forecasting method based on support vector regression machines which is superior to the time series forecasting method based on neural networks.