Identification of ISM Band Signals Using Deep Learning

Mingju He; Shengliang Peng; Huaxia Wang; Yu Dong Yao

doi:10.1109/WOCC48579.2020.9114911

Identification of ISM Band Signals Using Deep Learning

Mingju He, Shengliang Peng, Huaxia Wang, Yu Dong Yao

Department of Electrical and Computer Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

11 Scopus citations

Abstract

Spectrum awareness is now becoming more and more important in recent years, which can be utilized in areas like spectrum resource allocation, spectrum management, inference control, and security protection. Deep learning (DL) models, including convolutional neural network models have been widely used for classification related tasks, such as modulation classification, medium access control protocol (MAC) classification, and spectrum sensing. In this paper, a pre-trained Inception V3 model (CNN-based) is used to classify industrial, scientific, and medical (ISM) radio band signals. Experimentation results demonstrate the effectiveness of deep learning in ISM band signal identification.

Original language	English
Title of host publication	2020 29th Wireless and Optical Communications Conference, WOCC 2020
ISBN (Electronic)	9781728161242
DOIs	https://doi.org/10.1109/WOCC48579.2020.9114911
State	Published - May 2020
Event	29th Wireless and Optical Communications Conference, WOCC 2020 - Newark, United States Duration: 1 May 2020 → 2 May 2020

Publication series

Name	2020 29th Wireless and Optical Communications Conference, WOCC 2020

Conference

Conference	29th Wireless and Optical Communications Conference, WOCC 2020
Country/Territory	United States
City	Newark
Period	1/05/20 → 2/05/20

Access to Document

10.1109/WOCC48579.2020.9114911

Cite this

@inproceedings{be83856ced14487f8749961172d6e65d,

title = "Identification of ISM Band Signals Using Deep Learning",

abstract = "Spectrum awareness is now becoming more and more important in recent years, which can be utilized in areas like spectrum resource allocation, spectrum management, inference control, and security protection. Deep learning (DL) models, including convolutional neural network models have been widely used for classification related tasks, such as modulation classification, medium access control protocol (MAC) classification, and spectrum sensing. In this paper, a pre-trained Inception V3 model (CNN-based) is used to classify industrial, scientific, and medical (ISM) radio band signals. Experimentation results demonstrate the effectiveness of deep learning in ISM band signal identification.",

author = "Mingju He and Shengliang Peng and Huaxia Wang and Yao, {Yu Dong}",

note = "Publisher Copyright: {\textcopyright} 2020 IEEE.; 29th Wireless and Optical Communications Conference, WOCC 2020 ; Conference date: 01-05-2020 Through 02-05-2020",

year = "2020",

month = may,

doi = "10.1109/WOCC48579.2020.9114911",

language = "English",

series = "2020 29th Wireless and Optical Communications Conference, WOCC 2020",

booktitle = "2020 29th Wireless and Optical Communications Conference, WOCC 2020",

}

He, M, Peng, S, Wang, H & Yao, YD 2020, Identification of ISM Band Signals Using Deep Learning. in 2020 29th Wireless and Optical Communications Conference, WOCC 2020., 9114911, 2020 29th Wireless and Optical Communications Conference, WOCC 2020, 29th Wireless and Optical Communications Conference, WOCC 2020, Newark, United States, 1/05/20. https://doi.org/10.1109/WOCC48579.2020.9114911

TY - GEN

T1 - Identification of ISM Band Signals Using Deep Learning

AU - He, Mingju

AU - Peng, Shengliang

AU - Wang, Huaxia

AU - Yao, Yu Dong

PY - 2020/5

Y1 - 2020/5

N2 - Spectrum awareness is now becoming more and more important in recent years, which can be utilized in areas like spectrum resource allocation, spectrum management, inference control, and security protection. Deep learning (DL) models, including convolutional neural network models have been widely used for classification related tasks, such as modulation classification, medium access control protocol (MAC) classification, and spectrum sensing. In this paper, a pre-trained Inception V3 model (CNN-based) is used to classify industrial, scientific, and medical (ISM) radio band signals. Experimentation results demonstrate the effectiveness of deep learning in ISM band signal identification.

AB - Spectrum awareness is now becoming more and more important in recent years, which can be utilized in areas like spectrum resource allocation, spectrum management, inference control, and security protection. Deep learning (DL) models, including convolutional neural network models have been widely used for classification related tasks, such as modulation classification, medium access control protocol (MAC) classification, and spectrum sensing. In this paper, a pre-trained Inception V3 model (CNN-based) is used to classify industrial, scientific, and medical (ISM) radio band signals. Experimentation results demonstrate the effectiveness of deep learning in ISM band signal identification.

UR - http://www.scopus.com/inward/record.url?scp=85091927217&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85091927217&partnerID=8YFLogxK

U2 - 10.1109/WOCC48579.2020.9114911

DO - 10.1109/WOCC48579.2020.9114911

M3 - Conference contribution

AN - SCOPUS:85091927217

T3 - 2020 29th Wireless and Optical Communications Conference, WOCC 2020

BT - 2020 29th Wireless and Optical Communications Conference, WOCC 2020

T2 - 29th Wireless and Optical Communications Conference, WOCC 2020

Y2 - 1 May 2020 through 2 May 2020

ER -

Identification of ISM Band Signals Using Deep Learning

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this