Research And Application Of The Hilbert-huang Transform

In recent years, the theory and method of the Signal Processing have been developed rapidly. The analysis of the non-stationary signal is a new developing field in the modern Signal Processing. A new method, Hilbert-Huang Transform (HHT), was proposed by Huang et al. in 1998, which is especially developed for adaptively analyzing non-stationary and non- linear data. Since proposed, this method has already been applied to such a lot of non-linear research fields as turbulence, earthquake, finance, etc. successfully. In order to set up strict theory to this method, lots of work has to be done. In this paper, the application research field of HHT is extended and the extrema extending method of the upper and lower envelopes is developed.This paper is divided into four chapters altogether. Chapter 1 has introduced the background of the subject and rudimentary knowledge mainly.In Chapter 2, this paper has put forward a new extrema-extending algorithm in order to dispel the end wing phenomenon, and HHT is extended to image restoration field for the first time. By combining HHT and regularization, a new image restoration algorithm is presented. The extrema extending method of the upper and lower envelopes is developed. The selection of processing methods and regularization parameters depend on the feature of each intrinsic mode function. Some images with noise are processed with this algorithm. The experiment results show that the proposed algorithm is valid.The trading of stocks is a popular risk investment in modern business movement. But the stock market is unfathomable; investors have been hoping to control its variety regulation very much. In Chapter 3, a new stork operation model is presented by combining HHT. Intrinsic Mode Functions in the low frequency parts are got to replace the traditional motive average lines to analyze the data. The simulation results show that the proposed method gives the price differences better than those by traditional methods.The modal parameters identification is the important content of modal analysis. The Chapter 4 introduces the mathematical modal and structure modal response of Multi-degree of Freedom System. Then by combining HHT and state variable analysis, a method is proposed to identify modal frequency and damping ratio of time-invariant and time-varying multi-degree of system. The simulation results indicate that the time-frequency representation can be an effective tool for modal parameter identification of multi-degree of Freedom System.