Wavelets in compressing spectral data

Wavelet transformation has been developed during the last few years to improve the shortcomings of the Fourier transform technique. The major problem of Fourier transform is the complementarily of time and frequency resolution, i.e., when increasing the frequency resolution the transformation length increases and time resolution decreases considerably. In this paper some main aspects, related to the wavelet transform or wavelet analysis: such as denoising, smoothing, and compression multivariate data analysis. 1-D and 2-D spectral data is considered and illustrated by using, e.g., NIR and NMR data.