Minimax estimation of random vector in the presence of arbitrarily correlated disturbances

Аuthors

Lebedev M. V., Semenikhin K. V.^*

Moscow Aviation Institute (National Research University), 4, Volokolamskoe shosse, Moscow, А-80, GSP-3, 125993, Russia

*e-mail: siemenkv@rambler.ru

Abstract

A minimax estimation problem with mean-square criterion is discussed for stochastic observation model with white-noise and arbitrarily correlated disturbances. The least favorable covariance matrix is derived for the arbitrarily correlated disturbance. The minimax state estimate is obtained by means of this matrix for the analyzed system. The minimax algorithm is compared with other estimation techniques. Some results of numerical simulation are presented.