Accession Number : ADA254245
Title : Back-Propagation Network for Analog Signal Separation in High Noise Environments.
Descriptive Note : Final rept. Aug 91-Jan 92,
Corporate Author : CHEMICAL RESEARCH DEVELOPMENT AND ENGINEERING CENTER ABERDEEN PROVING GROUND MD
Personal Author(s) : Vanderbeek, Richard G ; Harper, Alice M
Report Date : Jul 1992
Pagination or Media Count : 15
Abstract : A back-propagation network is compared with principal components regression and prefiltered linear regression to demonstrate its ability to separate overlapped analog signals in high noise environments. Specifically, the signals tested were synthetically generated chemical mixture spectra that simulate the type of data obtained from chromatography and photo-spectrometry. The individual spectrum are heavily overlapped and 30 percent random noise and a random dc has been added to them. The comparisons were made for data sets comprised of two, three, and four overlapping spectrum. Neural networks, Chromatography, Back-propagation.
Descriptors : *NEURAL NETS , *MIXTURES , *CHROMATOGRAPHY , *ANALOG SIGNALS , PROPAGATION , ENVIRONMENTS , DETECTION , SIGNALS , FILTERS , NOISE , ANALOGS , SPECTRA , NETWORKS , CHEMICALS , COMPARISON
Subject Categories : Biochemistry
Distribution Statement : APPROVED FOR PUBLIC RELEASE