2014/02/01 by Mattia Zorzi, Zorzi, Mattia
Computer Science · Chemistry · #Blind Source Separation Techniques #Spectroscopy and Chemometric Analyses #Neural Networks and Applications
paper · pdf · doi:10.48550/arxiv.1402.0069
In this technical note, we deal with a spectrum approximation problem arising\nin THREE-like multivariate spectral estimation approaches. The solution to the\nproblem minimizes a suitable divergence index with respect to an a priori\nspectral density. We derive a new divergence family between multivariate\nspectral densities which takes root in the prediction theory. Under mild\nassumptions on the a priori spectral density, the approximation problem, based\non this new divergence family, admits a family of solutions. Moreover, an upper\nbound on the complexity degree of these solutions is provided.\n