Superresolution algorithms in discrete detectors and its testing for the silicon drift detectors data
Paper: 198
Session: A (poster)
Presenter: Kosarev, E.L., P.L.Kapitza Institute for Physical Problems, Moscow
Keywords: algorithms, analysis, data acquisition systems
Superresolution algorithms in discrete detectors and
its testing for the silicon drift detectors data
E.A.Kolganova, E.L.Kosarev and G.A.Ososkov
Joint Institute for Nuclear Research, Dubna, 141890
Moscow Region, Russia
P.L.Kapitza Institute for Physical Problems,
Russian Academy of Sciences, ul.Kosygina, 2, Moscow 117334, Russia
It is shown that the recovery of particle coordinates by detectors with
intrinsic resolution determined by the point spread function and finite
size of detector bins can be reduced to solution of the standard
deconvolution integral equation with modified point spread function. Two
approaches is proposed and investigated for this problem: parametric and
nonparametric ones. Algorithms based on maximum likelihood method [1]
and its testing for both approaches are given. It was shown that both
algorithms can resolve the coordinate of particles with resolving power
better than the size of detector bin and the point spread function
characteristic scale. It was also demonstrated that superresolution
efficiency of the parametric algorithm is attained the Cramer-Rao limit
and the Shannon's limit for the nonparametric algorithm [2]. Some
examples are presented for the silicon drift detector used at CERES/NA45.