Publication result detail

Improvements of Analog Neural Networks Based on Kalman Filter

TOBEŠ, Z., RAIDA, Z.

Original Title

Improvements of Analog Neural Networks Based on Kalman Filter

English Title

Improvements of Analog Neural Networks Based on Kalman Filter

Type

Peer-reviewed article not indexed in WoS or Scopus

Original Abstract

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.

English abstract

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.

Keywords

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

Key words in English

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

Authors

TOBEŠ, Z., RAIDA, Z.

Released

01.04.2002

ISBN

1210-2512

Periodical

Radioengineering

Volume

11

Number

3

State

Czech Republic

Pages from

6

Pages count

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"
}