MODEL OF OPTIMAL REGULATION OF NON-OBSERVABLE (HIDDEN) RANDOM PROCESS
Abstract
In the present paper the problem of non-observable (hidden) random autoregression process control based on the criteria of minimal dispersion of dynamical ratio value between an observable autoregression processes with hidden random process-dependent behavior is researched.
A sequence of controlled parameters is calculated for the specific case of problem. An approach reducing the calculation complexity of control parameter is found.
References
1. Kalman R.E. A New Approach to Linear Filtering and Prediction Problems// Transaction of the ASME-Jornal of Basic Engineering. – 1960. 82 (Series D): 35-45.
2. Панков А.Р., Миллер Г.Б. Минимаксная линейная рекуррентная фильтрация неопределенно-стохастических последовательностей по интегральному критерию// Информационные процессы. – 2001. – Т.1. - №2. – С.150-166
3. Agranovsky A.V. Lednov D.A. A model of controlled hidden linear autoregression processes for speakers identification// International Workshop ‘’SPEECH AND COMPUTER’’ (SPECOM). – 2005. – V2. – P.571-573.
Review
For citations:
LEDNOV D. MODEL OF OPTIMAL REGULATION OF NON-OBSERVABLE (HIDDEN) RANDOM PROCESS. Vestnik of Don State Technical University. 2006;6(2):125-133. (In Russ.)
Views:
331
Email this article 