A Gabor atom network for signal classification with application in radar target recognition

Author

Shi, Yu ; Zhang, Xian-Da

Author_Institution

Dept. of Autom., Tsinghua Univ., Beijing, China

Volume

49

Issue

12

fYear

2001

fDate

12/1/2001 12:00:00 AM

Firstpage

2994

Lastpage

3004

Abstract

A Gabor atom neural network approach is proposed for signal classification. The Gabor atom network uses a multilayer feedforward neural network structure, and its input layer constitutes the feature extraction part, whereas the hidden layer and the output layer constitute the signal classification part. From the physics point of view, it is shown that the time-shifted, frequency-modulated, and scaled Gaussian function is available for a basic model for the signal of high-resolution radar. Two experiment examples show that the Gabor atom network approach has a higher recognition rate in radar target recognition from range profiles as compared with several existing methods

Keywords

Gaussian processes; feature extraction; feedforward neural nets; radar computing; radar resolution; radar target recognition; signal classification; Gabor atom neural network; feature extraction; frequency modulation; hidden layer; high-resolution radar; input layer; multilayer feedforward neural network; output layer; radar target recognition; range profiles; scaled Gaussian function; signal classification; time-shifted function; Atomic layer deposition; Feature extraction; Feedforward neural networks; Frequency; Multi-layer neural network; Neural networks; Pattern classification; Physics; Radar; Target recognition;

fLanguage

English

Journal_Title

Signal Processing, IEEE Transactions on

Publisher

ieee

ISSN

1053-587X

Type

jour

DOI

10.1109/78.969508

Filename

969508