Solution of the inverse problem of light beating spectroscopy using singular value decomposition for analyzing the composition of polydisperse suspensions of nanoparticles

Abstract

The inverse problem of light beating spectroscopy is numerically solved by the singular value decomposition (SVD) method. The developed algorithm allows for the positivity of the solution and a constant background signal. Examples of reconstructed model uni-, bi-, and multimodal distributions are presented. The influences of the noise of a recorded spectrum, the background signal, and the chosen frequency range for measuring the signal on the characteristics of the distribution reconstructed by the SVD method are studied by the example of numerically generated (and measured with allowance for the noise) spectra.