Ulteriori informazioni
This book presents a general approach to block and recursive filtering, identification, and control, using signal observations processing techniques, and among others provides to the reader these results:
The new version of least square algorithm that is speeded up without changing its adaptive characteristics, increasing the parallelism in algorithm.
The efficient lower triangular inverse matrix and the input signal covariance matrix computation method.
The original bias correction approach that is used to eliminate the parameter estimation bias of an iterative autoregressive system parameter estimation algorithm in the presence of additive white noise.
The discovery that nonlinear Volterra, polynomial autoregressive and bilinear systems have the same layered implementation routine, which allows us using the layered structure, the order of nonlinearity increased/decreased by adding/deleting more layers to/from the structure.
The proven statement that the modular layered structures admit the very large scale integration implementation of the polynomial nonlinear filters.
The book is aimed at three major groups of readers: senior undergraduate students, graduate students, and scientific research workers in electrical engineering, computer engineering, computer science, and digital control.
Sommario
1. Introdution.- 2. High Speed Least Mean Square Adaptive Filtering.- 3. Optimal Parametri Identi ation of Linear Periodially Time-variant Systems.- 4. Iterative Parametric Identiation Algorithm of Autoregression.- 5. Layered Polynomial Filter Strutures.- 6. Exploring the Potentiality of Nonlinear Systems for Minimum Variane Control.- 7. Parametri Identiation of Systems with Pieewise-linear Nonlinearities.- 8. Parametri Identiation of Systems with Deadzones.- 9. Controlled Wiener System Simulation.- A Consistent parametri identiation.- Examples of the use of DSP methods for modelling, ltering, identiation, and control.
Info autore
In 1967, K. Kazlauskas finished Kaunas Polytechnical Institute (now Kaunas Technological University (KTU)) (Engineer of Technical Cybernetics). In 1973 at KTU, K. Kazlauskas defended PhD thesis (Cybernetics and Information Theory). In 1980 according the decision of Presidium of USSR Sciences, K. Kazlauskas was granted by the Senior Scientific Worker certificate in the field of Technical Cybernetics and Information Theory. In 1999 at Vytautas Magnus University (Kaunas) and Institute of Mathematics and Informatics (Vilnius), K. Kazlauskas defended Dr. Habilitus thesis ( Physical Sciences, Informatics). From 2000s according the decision of Vytautas Magnus University, K. Kazlauskas is a full professor (Physical Sciences, Informatics). K. Kazlauskas was a chief researcher and head of the group of Technological Processes Control at Institute of Mathematics and Informatics (1994–2015) and professor of Department of Informatics at Lithuanian University of Educology (2000–2017).
Rimantas Pupeikis finished Vilnius branch of Kaunas Politechnical Institute (now Kaunas Technological University (KTU)) in 1969s. He got a speciality of mechanization and automation of machines manufacture and acquired engineer's electromechanic's degree as well. In 1969s, R.Pupeikis, as junior scientific worker, began the scientific activity in the laboratory of adaptive systems of Institute of Energetics in Kaunas. In 1979s in KTU scientific board meeting, he defended his PhD thesis in Cybernetics and Information theory. In 1994s, he prepared the Doctor Habilitus thesis in Informatics. However, the governing body of Department of Informatics of Vytautas Magnum University (VDU) in Kaunas gave rise to his dissertation requirements that were not tuned up with official ones published by Lithuanian Sciences Council. Nevertheless, the dissertation left not defended because of absence of an official support.
Riassunto
This book presents a general approach to block and recursive filtering, identification, and control, using signal observations processing techniques, and among others provides to the reader these results:
The new version of least square algorithm that is speeded up without changing its adaptive characteristics, increasing the parallelism in algorithm.
The efficient lower triangular inverse matrix and the input signal covariance matrix computation method.
The original bias correction approach that is used to eliminate the parameter estimation bias of an iterative autoregressive system parameter estimation algorithm in the presence of additive white noise.
The discovery that nonlinear Volterra, polynomial autoregressive and bilinear systems have the same layered implementation routine, which allows us using the layered structure, the order of nonlinearity increased/decreased by adding/deleting more layers to/from the structure.
The proven statement that the modular layered structures admit the very large scale integration implementation of the polynomial nonlinear filters.
The book is aimed at three major groups of readers: senior undergraduate students, graduate students, and scientific research workers in electrical engineering, computer engineering, computer science, and digital control.
