ANN for real-time identification of radar objectives

Author

Bin Song ; Gong Long ; Junma Fu

Author_Institution

Dept. of Inf. & Control Eng., Xi´an Jiaotong Univ., Shaanxi, China

fYear

1992

fDate

18-25 June 1992

Firstpage

2277

Abstract

In order to complete the real-time identification of radar objectives with reliability, a mixed ANN (artificial neural network) that is made up of a function-link network and a modified BP (back propagation) network was introduced. The function-link network may increase the estimation speed, while the modified BP network that fully uses a parameter-estimating sequence ensures the reliability of identification. The experiment has been completed with the AST286 computer. The results show that this mixed network, which both ensures real time and realizes high reliability through a combined decision, is an effective way to accomplish the automatic and real-time identification of radar objectives.<>

Keywords

backpropagation; neural nets; parameter estimation; radar equipment; real-time systems; ANN; AST286 computer; artificial neural network; function-link network; modified backpropagation network; parameter-estimating sequence; radar objectives; real-time identification; reliability; Anisotropic magnetoresistance; Artificial neural networks; Control engineering; Data mining; Maximum likelihood estimation; Parameter estimation; Polarization; Radar applications; Shape; Smoothing methods;

fLanguage

English

Publisher

ieee

Conference_Titel

Antennas and Propagation Society International Symposium, 1992. AP-S. 1992 Digest. Held in Conjuction with: URSI Radio Science Meeting and Nuclear EMP Meeting., IEEE

Conference_Location

Chicago, IL, USA

Print_ISBN

0-7803-0730-5

Type

conf

DOI

10.1109/APS.1992.221397

Filename

221397