EFFICIENT COMPUTATION OF POLYNOMIAL SMOOTHING DIGITAL-FILTERS

Smoothing spectra with digital filters has many advantages over conventional RC fitters. The polynomial smoothing digital filters are especially suitable for spectrometry. This paper presents a method for reducing the computational requirements by several orders of magnitude by using a recursive algorithm. The method is stable and has zero numerical error while preserving all other properties of its well known nonrecursive ancestors. Formulas are given for the zero, second, and fourth degree polynomial filters and computational and storage requirements are discussed. © 1979, American Chemical Society. All rights reserved.