Title :
Recurrent filtering of non-Gaussian signals in natural conjugate class of distributions
Author_Institution :
St.-Petersburg State Univ. of Telecommun.
Abstract :
The problem of nonlinear filtering of stochastic signals on the basis of available non-Gaussian observations in discrete time is considered with applications to signal reception in channels with scattering (water, atmosphere). The method of recurrent filtering is proposed which is based on natural conjugate and minimax principles
Keywords :
electromagnetic wave scattering; filtering theory; minimax techniques; probability; signal detection; discrete time; distributions; minimax principles; natural conjugate class; non-Gaussian signals; recurrent filtering; scattering channels; signal reception; Atmosphere; Electronic mail; Equations; Filtering algorithms; Minimax techniques; Nonlinear filters; Parameter estimation; Probability density function; Radar scattering; Stochastic processes;
Conference_Titel :
Signal Processing Proceedings, 1998. ICSP '98. 1998 Fourth International Conference on
Conference_Location :
Beijing
Print_ISBN :
0-7803-4325-5
DOI :
10.1109/ICOSP.1998.770186
