An LSTM Approach to Short-range personnel recognition using Radar Signals

Zhenghui Li; Julien Le Kernec; Francesco Fioranelli; Olivier Romain; Lei  Zhang; Shufan Yang

doi:10.1109/RadarConf2147009.2021.9455218

An LSTM Approach to Short-range personnel recognition using Radar Signals

Zhenghui Li, Julien Le Kernec, Francesco Fioranelli, Olivier Romain, Lei Zhang, Shufan Yang

Microwave Sensing, Signals & Systems

Research output: Chapter in Book/Conference proceedings/Edited volume › Conference contribution › Scientific › peer-review

4 Citations (Scopus)

50 Downloads (Pure)

Abstract

In personnel recognition based on radar, significant research exists on statistical features extracted from the micro-Doppler signatures, whereas research considering other domains and information such as phase is less developed. This paper presents the use of deep learning methods to integrate both phase and magnitude features from range profiles and spectrogram. The temporal features of both domains are separately extracted using a stack of Long Short Term Memory (LSTM) layers. Then, the extracted features are aggregated in the corresponding domains and pass through a series of dense layers with SoftMax classifier. Finally, the information from the two domains is fused with a soft fusion approach to improve the performance further. Preliminary results show that the proposed network with soft fusion can achieve 85.5% accuracy in personnel recognition with six subjects

Original language	English
Title of host publication	2021 IEEE Radar Conference (RadarConf21)
Publisher	IEEE
Number of pages	6
ISBN (Electronic)	978-1-7281-7609-3
ISBN (Print)	978-1-7281-7610-9
DOIs	https://doi.org/10.1109/RadarConf2147009.2021.9455218
Publication status	Published - 2021
Event	2021 IEEE Radar Conference (RadarConf21): Radar on the Move - Atlanta, United States Duration: 7 May 2021 → 14 May 2021

Conference

Conference	2021 IEEE Radar Conference (RadarConf21)
Country/Territory	United States
City	Atlanta
Period	7/05/21 → 14/05/21

Bibliographical note

Green Open Access added to TU Delft Institutional Repository ‘You share, we take care!’ – Taverne project https://www.openaccess.nl/en/you-share-we-take-care

Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.

Keywords

Radar sensing
Personnel Recognition
LSTM network
Phase information
Micro-Doppler signatures
Range-time information

Access to Document

10.1109/RadarConf2147009.2021.9455218

09455218 (1)Final published version, 1.85 MB

Cite this

@inproceedings{57d1e77f17b1435aab426b61ec581f60,

title = "An LSTM Approach to Short-range personnel recognition using Radar Signals",

abstract = "In personnel recognition based on radar, significant research exists on statistical features extracted from the micro-Doppler signatures, whereas research considering other domains and information such as phase is less developed. This paper presents the use of deep learning methods to integrate both phase and magnitude features from range profiles and spectrogram. The temporal features of both domains are separately extracted using a stack of Long Short Term Memory (LSTM) layers. Then, the extracted features are aggregated in the corresponding domains and pass through a series of dense layers with SoftMax classifier. Finally, the information from the two domains is fused with a soft fusion approach to improve the performance further. Preliminary results show that the proposed network with soft fusion can achieve 85.5% accuracy in personnel recognition with six subjects",

keywords = "Radar sensing, Personnel Recognition, LSTM network, Phase information, Micro-Doppler signatures, Range-time information",

author = "Zhenghui Li and {Le Kernec}, Julien and Francesco Fioranelli and Olivier Romain and Lei Zhang and Shufan Yang",

note = "Green Open Access added to TU Delft Institutional Repository {\textquoteleft}You share, we take care!{\textquoteright} – Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.; 2021 IEEE Radar Conference (RadarConf21) : Radar on the Move ; Conference date: 07-05-2021 Through 14-05-2021",

year = "2021",

doi = "10.1109/RadarConf2147009.2021.9455218",

language = "English",

isbn = "978-1-7281-7610-9",

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

publisher = "IEEE",

address = "United States",

}

TY - GEN

T1 - An LSTM Approach to Short-range personnel recognition using Radar Signals

AU - Li, Zhenghui

AU - Le Kernec, Julien

AU - Fioranelli, Francesco

AU - Romain, Olivier

AU - Zhang, Lei

AU - Yang, Shufan

N1 - Green Open Access added to TU Delft Institutional Repository ‘You share, we take care!’ – Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.

PY - 2021

Y1 - 2021

N2 - In personnel recognition based on radar, significant research exists on statistical features extracted from the micro-Doppler signatures, whereas research considering other domains and information such as phase is less developed. This paper presents the use of deep learning methods to integrate both phase and magnitude features from range profiles and spectrogram. The temporal features of both domains are separately extracted using a stack of Long Short Term Memory (LSTM) layers. Then, the extracted features are aggregated in the corresponding domains and pass through a series of dense layers with SoftMax classifier. Finally, the information from the two domains is fused with a soft fusion approach to improve the performance further. Preliminary results show that the proposed network with soft fusion can achieve 85.5% accuracy in personnel recognition with six subjects

AB - In personnel recognition based on radar, significant research exists on statistical features extracted from the micro-Doppler signatures, whereas research considering other domains and information such as phase is less developed. This paper presents the use of deep learning methods to integrate both phase and magnitude features from range profiles and spectrogram. The temporal features of both domains are separately extracted using a stack of Long Short Term Memory (LSTM) layers. Then, the extracted features are aggregated in the corresponding domains and pass through a series of dense layers with SoftMax classifier. Finally, the information from the two domains is fused with a soft fusion approach to improve the performance further. Preliminary results show that the proposed network with soft fusion can achieve 85.5% accuracy in personnel recognition with six subjects

KW - Radar sensing

KW - Personnel Recognition

KW - LSTM network

KW - Phase information

KW - Micro-Doppler signatures

KW - Range-time information

UR - http://www.scopus.com/inward/record.url?scp=85112436895&partnerID=8YFLogxK

U2 - 10.1109/RadarConf2147009.2021.9455218

DO - 10.1109/RadarConf2147009.2021.9455218

M3 - Conference contribution

SN - 978-1-7281-7610-9

BT - 2021 IEEE Radar Conference (RadarConf21)

PB - IEEE

T2 - 2021 IEEE Radar Conference (RadarConf21)

Y2 - 7 May 2021 through 14 May 2021

ER -

An LSTM Approach to Short-range personnel recognition using Radar Signals

Abstract

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this