关键词: Hilbert-Huang变换, Hilbert变换, 端点效应, 经验模态分解, 支持矢量回归机

Abstract: 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 ma-chines are used to predict the signal before the signal is de-composed by EMD (Empirical Mode Decomposition), thus the end effects could be overcome effectively and the IMFs (Intrin-sic 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 ampli-tudes could be obtained and the Hilbert spectrum with physical sense could be acquired. The analysis results from the simu-lated 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 regres-sion machines which is superior to the time series forecasting method based on neural networks.

Key words: End effects Support vector regression machines Empirical mode decomposition Hilbert transform, Hilbert-Huang transform