Applied Spectroscopy · 1994 · 22 citations · 19 references
Natural SciencesChemistrySpectroscopic AnalysisSpectroscopyTime Series AnalysisBackground ReductionDigital Signal ProcessingRaman SpectroscopySignal ProcessingSpectral AnalysisSpectral Data ProcessingBackground NoiseBackground Noise ReductionFourier TransformSignal-to-noise AnalysisSingular Value DecompositionSmoothingSpectral Background SubtractionSpectral Smoothing
We propose a numerical method to eliminate the background noise from a line spectrum. The background noise is discriminated from signal components by applying a singular value decomposition (SVD) method to a cyclic matrix that is constructed from the spectrum data. The first column of the cyclic matrix is given by the Fourier transform of the spectrum data. Unlike conventional smoothing procedures, the new method gives no distortion to the signal waveform. The details of the method are described. Experimental results applied to FT-Raman spectrum data of carbon tetrachloride are also shown.
19
Partial least-squares regression: a tutorial
Paul Geladi, Bruce R. Kowalski · Analytica Chimica Acta · 1986 · 6.9K citations
Theory of error in factor analysis
Edmund R. Malinowski · Analytical Chemistry · 1977 · 263 citations
Factor analysis of the mass spectra of mixtures
G. L. Ritter, Stephen Lowry, T. L. Isenhour et al. · Analytical Chemistry · 1976 · 93 citations