Detail publikačního výsledku

Improvements of Analog Neural Networks Based on Kalman Filter

TOBEŠ, Z., RAIDA, Z.

Originální název

Improvements of Analog Neural Networks Based on Kalman Filter

Anglický název

Improvements of Analog Neural Networks Based on Kalman Filter

Druh

Článek recenzovaný mimo WoS a Scopus

Originální abstrakt

In the paper, original improvements of recurrent analog neural networks, which are based on Kalman filter, are presented. These improvements eliminate some disadvantages of the classical Kalman neural network and enable a real time processing of quickly changing signals, which appear in adaptive antennas and similar applications. This goal is reached using such circuit elements, which increase the convergence rate of the network and decrease the dependence of convergence rate on the ratio of eigenvalues of the correlation matrix of input signals.

Anglický abstrakt

In the paper, original improvements of recurrent analog neural networks, which are based on Kalman filter, are presented. These improvements eliminate some disadvantages of the classical Kalman neural network and enable a real time processing of quickly changing signals, which appear in adaptive antennas and similar applications. This goal is reached using such circuit elements, which increase the convergence rate of the network and decrease the dependence of convergence rate on the ratio of eigenvalues of the correlation matrix of input signals.

Klíčová slova

Kalman filter, analog recurrent neural networks, convergence rate, stability

Klíčová slova v angličtině

Kalman filter, analog recurrent neural networks, convergence rate, stability

Autoři

TOBEŠ, Z., RAIDA, Z.

Vydáno

01.04.2002

ISSN

1210-2512

Periodikum

Radioengineering

Svazek

11

Číslo

3

Stát

Česká republika

Strany od

6

Strany počet

8

BibTex

@article{BUT40903,
  author="Zdeněk {Tobeš} and Zbyněk {Raida}",
  title="Improvements of Analog Neural Networks Based on Kalman Filter",
  journal="Radioengineering",
  year="2002",
  volume="11",
  number="3",
  pages="8",
  issn="1210-2512"
}