Support Vector Machine based micro-Doppler signature classification of ground targets

Author

Javed, Azhar ; Liaqat, Sidrah ; bin Ihsan, Mojeeb

Author_Institution

Coll. of Electr. & Mech. Eng., Nat. Univ. of Sci. & Technol., Islamabad, Pakistan

fYear

2013

fDate

6-10 Oct. 2013

Firstpage

1827

Lastpage

1830

Abstract

In this paper, design of a micro-Doppler signature classifier for NR-V3 Ground Surveillance Radar* is discussed. The classifier distinguishes between pedestrians, vehicles and no target (noise) classes. Feature vector inputs for the classifier are extracted by preprocessing the FFT spectrum of radar backscattered signal. Support Vector Machine (SVM) with Radial Basis Function (RBF) and Polynomial kernels is used for classification of feature vectors. The classifiers are trained and tested using data collected with NR-V3 radar. This technique achieves a classification accuracy of over 94%.

Keywords

Doppler shift; fast Fourier transforms; feature extraction; polynomials; principal component analysis; radar computing; radar signal processing; radar target recognition; radial basis function networks; search radar; signal classification; support vector machines; FFT spectrum; NR-V3 ground surveillance radar; RBF; SVM; feature vector input classification; ground targets; microDoppler signature classification; pedestrians; polynomial kernels; principal component analysis; radar backscattered signal; radar target recognition; radial basis function; support vector machine; vehicles; Doppler radar; Feature extraction; Kernel; Support vector machine classification; Vehicles; Automatic target classification; Ground surveillance radar; Principal component analysis; Pulsedoppler radar; Radar target recognition; Support Vector Machine; micro-Doppler;

fLanguage

English

Publisher

ieee

Conference_Titel

Microwave Conference (EuMC), 2013 European

Conference_Location

Nuremberg

Type

conf

Filename

6687035