The Identification of Image Signals by Robust Systems with Stochastic
Pumping
V.
I.
Nevolin,
M.
B.
Merenkov,
D.
V.
Ivaneev,
M.
A.
Zuikova
Federal State
State-Financed Educational Institution of High Professional
Education “South Ural State University” (national research university)
Received
January 16, 2013
Abstract:
The problems of methodology of robust
systems of the identification of signals with unknown parameters as well as
the identification of image signals under the conditions of powerful
additive hindrances, e.g. for space radiosonde observation, are dealt with
in the article. The nonparametric stochastic approach based on selective
distributions of observations is used here. The synthesis of the schemes of
robust filters is carried out by the robust nonlinear filtration method, and
the identification (filtration and detection) of signals – by constructing
robust partial systems. The basis of physical increase of robust systems’
noise stability due to the phenomenon of power transformations under the
conditions of stochastic automatic rating from attendant additive hindrances
is laid down in the paper. The results of the conducted simulation show the
essential increase of noise stability in comparison with the coordinated
filtration.
Keywords: robust methods, nonlinear filtering, numerical modeling, stochastic pumping.