Interference motion removal for Doppler radar vital sign detection using variational encoder-decoder neural network

Mikolaj Czerkawski; Christos Ilioudis; Carmine Clemente; Craig Michie; Ivan Andonovic; Christos Tachtatzis

doi:10.1109/RadarConf2147009.2021.9454986

Interference motion removal for Doppler radar vital sign detection using variational encoder-decoder neural network

Mikolaj Czerkawski, Christos Ilioudis, Carmine Clemente, Craig Michie, Ivan Andonovic, Christos Tachtatzis

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution book

7 Citations (Scopus)

147 Downloads (Pure)

Abstract

The treatment of interfering motion contributions remains one of the key challenges in the domain of radar-based vital sign monitoring. Removal of the interference to extract the vital sign contributions is demanding due to overlapping Doppler bands, the complex structure of the interference motions and significant variations in the power levels of their contributions. A novel approach to the removal of interference through the use of a probabilistic deep learning model is presented. Results show that a convolutional encoder-decoder neural network with a variational objective is capable of learning a meaningful representation space of vital sign Doppler-time distribution facilitating their extraction from a mixture signal. The approach is tested on semi-experimental data containing real vital sign signatures and simulated returns from interfering body motions. It is demonstrated that the application of the proposed network enhances the extraction of the micro-Doppler frequency corresponding to the respiration rate.

Original language	English
Title of host publication	2021 IEEE Radar Conference (RadarConf21)
Place of Publication	New York, NY.
Publisher	IEEE
Pages	1-6
Number of pages	6
ISBN (Electronic)	9781728176093
ISBN (Print)	9781728176109
DOIs	https://doi.org/10.1109/RadarConf2147009.2021.9454986
Publication status	Published - 14 May 2021
Event	2021 IEEE Radar Conference - Virtual/Atlanta, GA, USA, Atlanta, United States Duration: 10 May 2021 → 14 May 2021 https://ewh.ieee.org/conf/radar/2021/

Publication series

Name	IEEE National Radar Conference - Proceedings
Volume	2021-May
ISSN (Print)	1097-5659

Conference

Conference	2021 IEEE Radar Conference
Abbreviated title	RadarConf 2021
Country/Territory	United States
City	Atlanta
Period	10/05/21 → 14/05/21
Internet address	https://ewh.ieee.org/conf/radar/2021/

Keywords

Doppler radar
heart rate monitoring
respiration rate monitoring
vital signs
random body movement
variational autoencoder

Access to Document

10.1109/RadarConf2147009.2021.9454986

Czerkawski-etal-IEEE-RadarConf2021-Interference-motion-removal-for-Doppler-radar-vital-sign-detectionAccepted author manuscript, 1.35 MB

Doctoral Training Partnership 2018-19 University of Strathclyde | Czerkawski, Mikolaj
Tachtatzis, C. (Principal Investigator), Clemente, C. (Co-investigator) & Czerkawski, M. (Research Co-investigator)
EPSRC (Engineering and Physical Sciences Research Council)
1/10/19 → 11/08/23
Project: Research Studentship - Internally Allocated
KTP - Fathom Systems
Tachtatzis, C. (Principal Investigator), Andonovic, I. (Co-investigator), Atkinson, R. (Co-investigator), Cleary, A. (Co-investigator) & Michie, C. (Co-investigator)
Fathom Systems Limited, KTP Govt (TSB)
4/09/17 → 3/09/20
Project: Research

Human Walking Motion Radar Simulator
Czerkawski, M. (Creator), Clemente, C. (Supervisor) & Tachtatzis, C. (Supervisor), Zenodo, 2 Feb 2021
DOI: 10.5281/zenodo.4245158, https://github.com/cidcom/human-walking-radar-simulator/tree/0.1.1
Dataset

Cite this

Czerkawski, M., Ilioudis, C., Clemente, C., Michie, C., Andonovic, I., & Tachtatzis, C. (2021). Interference motion removal for Doppler radar vital sign detection using variational encoder-decoder neural network. In 2021 IEEE Radar Conference (RadarConf21) (pp. 1-6). (IEEE National Radar Conference - Proceedings; Vol. 2021-May). IEEE. https://doi.org/10.1109/RadarConf2147009.2021.9454986

@inproceedings{5a43196db9d54a46b7be0b7c9a553584,

title = "Interference motion removal for Doppler radar vital sign detection using variational encoder-decoder neural network",

abstract = "The treatment of interfering motion contributions remains one of the key challenges in the domain of radar-based vital sign monitoring. Removal of the interference to extract the vital sign contributions is demanding due to overlapping Doppler bands, the complex structure of the interference motions and significant variations in the power levels of their contributions. A novel approach to the removal of interference through the use of a probabilistic deep learning model is presented. Results show that a convolutional encoder-decoder neural network with a variational objective is capable of learning a meaningful representation space of vital sign Doppler-time distribution facilitating their extraction from a mixture signal. The approach is tested on semi-experimental data containing real vital sign signatures and simulated returns from interfering body motions. It is demonstrated that the application of the proposed network enhances the extraction of the micro-Doppler frequency corresponding to the respiration rate.",

keywords = "Doppler radar, heart rate monitoring, respiration rate monitoring, vital signs, random body movement, variational autoencoder",

author = "Mikolaj Czerkawski and Christos Ilioudis and Carmine Clemente and Craig Michie and Ivan Andonovic and Christos Tachtatzis",

note = "{\textcopyright} 2021 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting /republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.; 2021 IEEE Radar Conference, RadarConf 2021 ; Conference date: 10-05-2021 Through 14-05-2021",

year = "2021",

month = may,

day = "14",

doi = "10.1109/RadarConf2147009.2021.9454986",

language = "English",

isbn = "9781728176109",

series = "IEEE National Radar Conference - Proceedings",

publisher = "IEEE",

pages = "1--6",

booktitle = "2021 IEEE Radar Conference (RadarConf21)",

url = "https://ewh.ieee.org/conf/radar/2021/",

}

Czerkawski, M, Ilioudis, C, Clemente, C , Michie, C , Andonovic, I & Tachtatzis, C 2021, Interference motion removal for Doppler radar vital sign detection using variational encoder-decoder neural network. in 2021 IEEE Radar Conference (RadarConf21). IEEE National Radar Conference - Proceedings, vol. 2021-May, IEEE, New York, NY., pp. 1-6, 2021 IEEE Radar Conference, Atlanta, United States, 10/05/21. https://doi.org/10.1109/RadarConf2147009.2021.9454986

Interference motion removal for Doppler radar vital sign detection using variational encoder-decoder neural network. / Czerkawski, Mikolaj; Ilioudis, Christos; Clemente, Carmine et al.
2021 IEEE Radar Conference (RadarConf21). New York, NY.: IEEE, 2021. p. 1-6 (IEEE National Radar Conference - Proceedings; Vol. 2021-May).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution book

TY - GEN

T1 - Interference motion removal for Doppler radar vital sign detection using variational encoder-decoder neural network

AU - Czerkawski, Mikolaj

AU - Ilioudis, Christos

AU - Clemente, Carmine

AU - Michie, Craig

AU - Andonovic, Ivan

AU - Tachtatzis, Christos

N1 - © 2021 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting /republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.

PY - 2021/5/14

Y1 - 2021/5/14

N2 - The treatment of interfering motion contributions remains one of the key challenges in the domain of radar-based vital sign monitoring. Removal of the interference to extract the vital sign contributions is demanding due to overlapping Doppler bands, the complex structure of the interference motions and significant variations in the power levels of their contributions. A novel approach to the removal of interference through the use of a probabilistic deep learning model is presented. Results show that a convolutional encoder-decoder neural network with a variational objective is capable of learning a meaningful representation space of vital sign Doppler-time distribution facilitating their extraction from a mixture signal. The approach is tested on semi-experimental data containing real vital sign signatures and simulated returns from interfering body motions. It is demonstrated that the application of the proposed network enhances the extraction of the micro-Doppler frequency corresponding to the respiration rate.

AB - The treatment of interfering motion contributions remains one of the key challenges in the domain of radar-based vital sign monitoring. Removal of the interference to extract the vital sign contributions is demanding due to overlapping Doppler bands, the complex structure of the interference motions and significant variations in the power levels of their contributions. A novel approach to the removal of interference through the use of a probabilistic deep learning model is presented. Results show that a convolutional encoder-decoder neural network with a variational objective is capable of learning a meaningful representation space of vital sign Doppler-time distribution facilitating their extraction from a mixture signal. The approach is tested on semi-experimental data containing real vital sign signatures and simulated returns from interfering body motions. It is demonstrated that the application of the proposed network enhances the extraction of the micro-Doppler frequency corresponding to the respiration rate.

KW - Doppler radar

KW - heart rate monitoring

KW - respiration rate monitoring

KW - vital signs

KW - random body movement

KW - variational autoencoder

U2 - 10.1109/RadarConf2147009.2021.9454986

DO - 10.1109/RadarConf2147009.2021.9454986

M3 - Conference contribution book

SN - 9781728176109

T3 - IEEE National Radar Conference - Proceedings

SP - 1

EP - 6

BT - 2021 IEEE Radar Conference (RadarConf21)

PB - IEEE

CY - New York, NY.

T2 - 2021 IEEE Radar Conference

Y2 - 10 May 2021 through 14 May 2021

ER -

Interference motion removal for Doppler radar vital sign detection using variational encoder-decoder neural network

Abstract

Publication series

Conference

Keywords

Access to Document

Fingerprint

Projects

Doctoral Training Partnership 2018-19 University of Strathclyde | Czerkawski, Mikolaj

KTP - Fathom Systems

Datasets

Human Walking Motion Radar Simulator

Cite this