Infrared spectroscopy is useful to monitor the quality of products on-line, or to quality multivariate properties simultaneously. The IR spectrometer satisfies the requirements of users who want to have quantitative product information in real-time because the instrument provides the information promptly and easily. However, Samples that are measured using diffuse reflectance often exhibit significant differences in the spectra due to the non-homogeneous distribution of the particles. In fact, multiple spectral measurements of the same sample can look completely different. In many cases, the scattering can be an overpowering contributor to the spectrum, sometimes accounting for most of the variance in the data. Although the degree of scattering is dependent on the wavelength of the light that is used and the particle size and refractive index of the sample, the scattering is not uniform throughout the spectrum. Typically, this appears as a baseline shift, tilt and sometimes curvature, where the degree of influence is more pronounced at the longer-wavelength end of the spectrum. The diffuse reflection spectrum is unsatisfactory and the calibration may provide unsatisfactory prediction results. So we must use some methods to remove the effects of the scattering for multivariate calibration of IR spectral signals. Discrete wavelet transform (DWT) is a good method to remove the effects of the scattering for multivariate calibration of IR spectral signals. By using DWT on individual signals as a preprocessing method in regression modeling on IR spectra, good compression is achieved with almost no loss of information, the low-frequency varying background and the high-frequency noise be removed simultaneously. In this report, we use the iPLS method to establish the calibration models of hydrocarbon source rocks. iPLS is a new regression method and the authors can get better results by using DTW- iPLS.