Improving Identification Algorithm for Mixed Signal Systems
System Identification and Control Problems: VI International Conference (SICPRO 07) 2007
Valerijs Zagurskis

In this paper we propose an verification (identification) procedure applicable to mixed signal non-linear objects if structure of them are close to the Hammerstein model. Decomposition of the static characteristic is used to get a simple and common approach for a wide class of non-essential non-linear ties and it allows calculating the output signal for any input signal. To deal with the dynamics, coefficients of the differential equation arc used. As a result, the behavior of the object can be fully described using only a small set of parameters, and each of them can be estimated experimentally. The simulation results confirmed the usefulness of the model being proposed.


Keywords
Identification, mixed signal systems
Hyperlink
www.sicpro.org

Zagurskis, V. Improving Identification Algorithm for Mixed Signal Systems. In: System Identification and Control Problems: VI International Conference (SICPRO 07), Russia, Moscow, 29 Jan-3 Feb., 2007. Moscow: RAN Institute of Control Sciences, 2007, pp.937-942.

Publication language
English (en)
The Scientific Library of the Riga Technical University.
E-mail: uzzinas@rtu.lv; Phone: +371 28399196