Time series recognition based on wavelet transform and Fourier transform

Time series classification based on wavelet transforms and Fourier transform is discussed in this paper. Wavelet transforms have the time-variant characteristic, and are relatively sensitive to the time series with some mutations. Fourier transform is able to reflect various periodic variation of time series clearly. The test proves that the hierarchical clustering based on wavelet transforms can fully manifest the subtle differences among time series, while the hierarchical clustering based on Fourier transform may classify time series from the overall perspective.