The Application And Extension Of Chirplet Transform

The varying time-frequency relation of sedimentary cycle signals reflects the tendency for the surface to change from thick to thin or vice verse. It has the practical significance in search for the subtle reservoirs , such as complicated lithology, thin interbeds and so on. In order to obtain the important information as well as its features in the records, people traditionally tend to adopt Fourier Transform, which only indicates the amount of frequency in the signal instead of reflecting the time-based variations of the frequency. Recently, with the development of information technology, various signal processing technologies of checking the varying time-frequency relation emerges and flourishes. These technologies mainly include: Short-time Fourier Transforms (STFT), Wavelet Transforms (WAVELET TRANSFORM), Wigner-Ville Distribution. The thesis further probes the application of the three methods of signal processing to the instantaneous frequency computation , identifies their features and compares the advantages and disadvantages of their respective applications. Further study, based on the above issues, has been carried on concerning the newly-appeared Chirplet Transform in recent years, with the experiment attached to explain the advantages in analyzing cycle signals.The main contributions of this thesis are as follows:(一) The theoretic system of STFT, Wavelet Transforms, and Wigner-Ville are thoroughly discussed, including the basic theory, the basis and the common features. Through the time-frequency analysis of the signals, we can primarily infer the divergence in presenting content as well as their respective advantages and disadvantages.(二) The detailed analysis of the time-frequency relation from the theoretical perspective is provided, which lays the foundation for the further study, and the common methods of estimating the instantaneous frequency of the signals are compared as well.