A method is proposed for the determination of the optimum value of the regularization parameter (Lagrange multiplier) when applying indirect transform techniques in small-angle scattering data analysis. The method is based on perceptual criteria of what is the best solution. A set of simple criteria is used to construct a total estimate describing the quality of the solution. Maximization of the total estimate is straightforward. Model computations show the effectiveness of the technique.
A program suite for one-dimensional small-angle scattering data processing running on IBM-compatible PCs under Windows 9x/NT/2000/XP is presented. The main program, PRIMUS, has a menu-driven graphical user interface calling computational modules to perform data manipulation and analysis. Experimental data in binary OTOKO format can be reduced by calling the program SAPOKO, which includes statistical analysis of time frames, averaging and scaling. Tools to generate the angular axis and detector response ®les from diffraction patterns of calibration samples, as well as binary to ASCII transformation programs, are available. Several types of ASCII ®les can be directly imported into PRIMUS, in particular, sasCIF or ILL-type ®les are read without modi®cation. PRIMUS provides basic data manipulation functions (averaging, background subtraction, merging of data measured in different angular ranges, extrapolation to zero sample concentration, etc.) and computes invariants from Guinier and Porod plots. Several external modules coupled with PRIMUS via pop-up menus enable the user to evaluate the characteristic functions by indirect Fourier transformation, to perform peak analysis for partially ordered systems and to ®nd shape approximations in terms of threeparametric geometrical bodies. For the analysis of mixtures, PRIMUS enables model-independent singular value decomposition or linear ®tting if the scattering from the components is known. An interface is also provided to the general non-linear ®tting program MIXTURE, which is designed for quantitative analysis of multicomponent systems represented by simple geometrical bodies, taking shape and size polydispersity as well as interparticle interference effects into account.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.