Radar classification of indoor targets using support vector machines

Research output: Contribution to journalArticle

12 Citations (Scopus)

Abstract

Through-the-wall radar systems are typically plagued by harsh clutter which confounds the detection and classification of targets. This study approaches the problem of separating humans from indoor targets through machine learning techniques by implementing support vector machine (SVM) classifiers. Simulated and experimental data are used to evaluate the suitability of SVMs for target classification via their spectral characteristics. Numerical modelling of humans and indoor clutter was accomplished through the finite-difference time-domain (FDTD) method. The radar cross-sections for the various objects are acquired over a wide range of frequencies, polarisations, aspect angles, and material properties. Furthermore, the spectral properties of humans and clutter targets are measured using a network analyser. The potentials of the acquired simulated and experimental spectral characteristics at different frequencies and polarisations are explored for target classification using SVMs. Feature selection algorithms are also investigated to reduce the redundancy and model complexity. Finally, the classification performance is assessed in the presence of additive white Gaussian noise and through various wall materials.

Original languageEnglish (US)
Pages (from-to)1468-1476
Number of pages9
JournalIET Radar, Sonar and Navigation
Volume10
Issue number8
DOIs
StatePublished - Jan 1 2016

Fingerprint

Support vector machines
Radar
Polarization
Radar cross section
Finite difference time domain method
Radar systems
Redundancy
Learning systems
Feature extraction
Materials properties
Classifiers

All Science Journal Classification (ASJC) codes

  • Electrical and Electronic Engineering

Cite this

@article{b7a974104f024cda9ab136ce373e7013,
title = "Radar classification of indoor targets using support vector machines",
abstract = "Through-the-wall radar systems are typically plagued by harsh clutter which confounds the detection and classification of targets. This study approaches the problem of separating humans from indoor targets through machine learning techniques by implementing support vector machine (SVM) classifiers. Simulated and experimental data are used to evaluate the suitability of SVMs for target classification via their spectral characteristics. Numerical modelling of humans and indoor clutter was accomplished through the finite-difference time-domain (FDTD) method. The radar cross-sections for the various objects are acquired over a wide range of frequencies, polarisations, aspect angles, and material properties. Furthermore, the spectral properties of humans and clutter targets are measured using a network analyser. The potentials of the acquired simulated and experimental spectral characteristics at different frequencies and polarisations are explored for target classification using SVMs. Feature selection algorithms are also investigated to reduce the redundancy and model complexity. Finally, the classification performance is assessed in the presence of additive white Gaussian noise and through various wall materials.",
author = "Bufler, {Travis Dale} and Narayanan, {Ram Mohan}",
year = "2016",
month = "1",
day = "1",
doi = "10.1049/iet-rsn.2015.0580",
language = "English (US)",
volume = "10",
pages = "1468--1476",
journal = "IET Radar, Sonar and Navigation",
issn = "1751-8784",
publisher = "Institution of Engineering and Technology",
number = "8",

}

Radar classification of indoor targets using support vector machines. / Bufler, Travis Dale; Narayanan, Ram Mohan.

In: IET Radar, Sonar and Navigation, Vol. 10, No. 8, 01.01.2016, p. 1468-1476.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Radar classification of indoor targets using support vector machines

AU - Bufler, Travis Dale

AU - Narayanan, Ram Mohan

PY - 2016/1/1

Y1 - 2016/1/1

N2 - Through-the-wall radar systems are typically plagued by harsh clutter which confounds the detection and classification of targets. This study approaches the problem of separating humans from indoor targets through machine learning techniques by implementing support vector machine (SVM) classifiers. Simulated and experimental data are used to evaluate the suitability of SVMs for target classification via their spectral characteristics. Numerical modelling of humans and indoor clutter was accomplished through the finite-difference time-domain (FDTD) method. The radar cross-sections for the various objects are acquired over a wide range of frequencies, polarisations, aspect angles, and material properties. Furthermore, the spectral properties of humans and clutter targets are measured using a network analyser. The potentials of the acquired simulated and experimental spectral characteristics at different frequencies and polarisations are explored for target classification using SVMs. Feature selection algorithms are also investigated to reduce the redundancy and model complexity. Finally, the classification performance is assessed in the presence of additive white Gaussian noise and through various wall materials.

AB - Through-the-wall radar systems are typically plagued by harsh clutter which confounds the detection and classification of targets. This study approaches the problem of separating humans from indoor targets through machine learning techniques by implementing support vector machine (SVM) classifiers. Simulated and experimental data are used to evaluate the suitability of SVMs for target classification via their spectral characteristics. Numerical modelling of humans and indoor clutter was accomplished through the finite-difference time-domain (FDTD) method. The radar cross-sections for the various objects are acquired over a wide range of frequencies, polarisations, aspect angles, and material properties. Furthermore, the spectral properties of humans and clutter targets are measured using a network analyser. The potentials of the acquired simulated and experimental spectral characteristics at different frequencies and polarisations are explored for target classification using SVMs. Feature selection algorithms are also investigated to reduce the redundancy and model complexity. Finally, the classification performance is assessed in the presence of additive white Gaussian noise and through various wall materials.

UR - http://www.scopus.com/inward/record.url?scp=84982253251&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84982253251&partnerID=8YFLogxK

U2 - 10.1049/iet-rsn.2015.0580

DO - 10.1049/iet-rsn.2015.0580

M3 - Article

VL - 10

SP - 1468

EP - 1476

JO - IET Radar, Sonar and Navigation

JF - IET Radar, Sonar and Navigation

SN - 1751-8784

IS - 8

ER -