IEICE TRANSACTIONS on Communications
Classification of Electromagnetic Radiation Source Models Based on Directivity with the Method of Machine Learning
Summary :
This paper presents a new way to classify different radiation sources by the parameter of directivity, which is a characteristic parameter of electromagnetic radiation sources. The parameter can be determined from measurements of the electric field intensity radiating in all directions in space. We develop three basic antenna models, which are for 3GHz operation, and set 125,000 groups of cube receiving arrays along the main lobe of their radiation patterns to receive the data of far field electric intensity in groups. Then the Back Propagation (BP) neural network and the Support Vector Machine (SVM) method are adopted to analyze training data set, and build and test the classification model. Owing to the powerful nonlinear simulation ability, the SVM method offers higher classification accuracy than the BP neural network in noise environment. At last, the classification model is comprehensively evaluated in three aspects, which are capability of noise immunity, F1 measure and the normalization method.
- Publication
- IEICE TRANSACTIONS on Communications Vol.E98-B No.7 pp.1227-1234
- Publication Date
- 2015/07/01
- Publicized
- Online ISSN
- 1745-1345
- DOI
- 10.1587/transcom.E98.B.1227
- Type of Manuscript
- Special Section PAPER (Special Section on Electromagnetic Compatibility Technology in Conjunction with Main Topics of EMC'14/Tokyo)
- Category
Authors
Zhuo LIU
Beijing University of Posts and Telecommunications
Dan SHI
Beijing University of Posts and Telecommunications
Yougang GAO
Beijing University of Posts and Telecommunications
Junjian BI
Shijiazhuang Mech. Eng. Coll.
Zhiliang TAN
Shijiazhuang Mech. Eng. Coll.
Jingjing SHI
Nagoya Institute of Technology
Keyword
Latest Issue
Copyrights notice of machine-translated contents
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Cite this
Copy
Zhuo LIU, Dan SHI, Yougang GAO, Junjian BI, Zhiliang TAN, Jingjing SHI, "Classification of Electromagnetic Radiation Source Models Based on Directivity with the Method of Machine Learning" in IEICE TRANSACTIONS on Communications,
vol. E98-B, no. 7, pp. 1227-1234, July 2015, doi: 10.1587/transcom.E98.B.1227.
Abstract: This paper presents a new way to classify different radiation sources by the parameter of directivity, which is a characteristic parameter of electromagnetic radiation sources. The parameter can be determined from measurements of the electric field intensity radiating in all directions in space. We develop three basic antenna models, which are for 3GHz operation, and set 125,000 groups of cube receiving arrays along the main lobe of their radiation patterns to receive the data of far field electric intensity in groups. Then the Back Propagation (BP) neural network and the Support Vector Machine (SVM) method are adopted to analyze training data set, and build and test the classification model. Owing to the powerful nonlinear simulation ability, the SVM method offers higher classification accuracy than the BP neural network in noise environment. At last, the classification model is comprehensively evaluated in three aspects, which are capability of noise immunity, F1 measure and the normalization method.
URL: https://global.ieice.org/en_transactions/communications/10.1587/transcom.E98.B.1227/_p
Copy
@ARTICLE{e98-b_7_1227,
author={Zhuo LIU, Dan SHI, Yougang GAO, Junjian BI, Zhiliang TAN, Jingjing SHI, },
journal={IEICE TRANSACTIONS on Communications},
title={Classification of Electromagnetic Radiation Source Models Based on Directivity with the Method of Machine Learning},
year={2015},
volume={E98-B},
number={7},
pages={1227-1234},
abstract={This paper presents a new way to classify different radiation sources by the parameter of directivity, which is a characteristic parameter of electromagnetic radiation sources. The parameter can be determined from measurements of the electric field intensity radiating in all directions in space. We develop three basic antenna models, which are for 3GHz operation, and set 125,000 groups of cube receiving arrays along the main lobe of their radiation patterns to receive the data of far field electric intensity in groups. Then the Back Propagation (BP) neural network and the Support Vector Machine (SVM) method are adopted to analyze training data set, and build and test the classification model. Owing to the powerful nonlinear simulation ability, the SVM method offers higher classification accuracy than the BP neural network in noise environment. At last, the classification model is comprehensively evaluated in three aspects, which are capability of noise immunity, F1 measure and the normalization method.},
keywords={},
doi={10.1587/transcom.E98.B.1227},
ISSN={1745-1345},
month={July},}
Copy
TY - JOUR
TI - Classification of Electromagnetic Radiation Source Models Based on Directivity with the Method of Machine Learning
T2 - IEICE TRANSACTIONS on Communications
SP - 1227
EP - 1234
AU - Zhuo LIU
AU - Dan SHI
AU - Yougang GAO
AU - Junjian BI
AU - Zhiliang TAN
AU - Jingjing SHI
PY - 2015
DO - 10.1587/transcom.E98.B.1227
JO - IEICE TRANSACTIONS on Communications
SN - 1745-1345
VL - E98-B
IS - 7
JA - IEICE TRANSACTIONS on Communications
Y1 - July 2015
AB - This paper presents a new way to classify different radiation sources by the parameter of directivity, which is a characteristic parameter of electromagnetic radiation sources. The parameter can be determined from measurements of the electric field intensity radiating in all directions in space. We develop three basic antenna models, which are for 3GHz operation, and set 125,000 groups of cube receiving arrays along the main lobe of their radiation patterns to receive the data of far field electric intensity in groups. Then the Back Propagation (BP) neural network and the Support Vector Machine (SVM) method are adopted to analyze training data set, and build and test the classification model. Owing to the powerful nonlinear simulation ability, the SVM method offers higher classification accuracy than the BP neural network in noise environment. At last, the classification model is comprehensively evaluated in three aspects, which are capability of noise immunity, F1 measure and the normalization method.
ER -
English
