Wavelet Transform In Signal Processing - Sound And Vibration Signal Analysis And Digital Watermarking Technology

As a new theory of time-frequency domain, Wavelet Transform is seen as the new phase of Fourier Transform. It not only inherits and develops STFT localize, but also overcomes shortcomings, such as window size not changing with the frequency. Based on the merits, the Wavelet Transform is broadly used in various signals processing.The paper puts the emphasis on the signal de-noising and calculation time-difference for collected sound wave and vibration signals. At the same time, the Wavelet Transform is applied to embed and detect watermark.Aiming at various signals de-noising, various wavelet transform methods are applied: (l)de-nosing the higher SNR signals collected on the earth, comparing the Wavelet Transform thresholding with the horizontal motion invariable Wavelet Transform, the paper concludes that the latter is more fitter to de-noise the kind of signals;(2) de-noising the lower SNR signals collected in the water, the paper uses the Wavelet Transform power spectrum to de-noise the signals and achieves the better results.Aiming at the time-difference of the same source signals, the wavelet characteristic of singularity detection is applied.In this paper, the wavelet decomposition and re-composition are applied to the watermark embedding and detection.