EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A Variational Autoencoder Network for Heartbeat Signal Reconstruction and Breathing-State Classification Using 24 GHz Radar Sensors

Homa Arab iD, Nima Suzandeh, Serioja O Tatu iD, Steven Dufour iD

Sensors · 09.11.2026 · 26 (21) · 7083

DOI: 10.3390/s26217083 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 09.11.2026. Verfasst von Homa Arab und 3 weiteren Autor:innen. Beteiligte Einrichtung: Department of Mathematics and Industrial Engineering, Polytechnique Montreal, Montreal, QC H3T 1J4, Canada. Hauptthemen laut MEDLINE: Signal Processing, Computer-Assisted, Heart Rate, Respiration, Radar. Die Arbeit nennt 28 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Wireless radar sensors can remotely detect vital signs. This study investigates the wireless monitoring of physiological signals induced by respiration and heartbeat to classify breathing states. Two 24 GHz radar sensors are used. Noise, harmonics, and intermodulation products affect the measured physiological signals and can make heartbeat identification and analysis challenging. Deep-learning techniques have recently been used to denoise and classify signals in wireless communications. Variational autoencoders (VAEs) have also been used for signal and image processing. Multilayer VAEs support sequential learning by optimizing generative-model parameters and reducing reconstruction error. This study investigates the separation of measured heartbeat and respiratory signals and the reduction in noise and the effects of limited random body movements using VAEs under controlled conditions. Two radars are positioned in front of and behind a human subject to record vital signals simultaneously. The separated and denoised heartbeat signals are input to a ResNet network to classify normal and fast breathing states. The proposed method achieves 99.7% accuracy on the reported experimental dataset.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Signal Processing, Computer-AssistedHeart RateRespirationRadarAutoencoderHumansAlgorithmsConvolutional Neural NetworksWireless TechnologyNeural Networks, Computer

Stichwörter der Autor:innen

Doppler frequencyResNetartificial intelligenceconvolutional neural network (CNN)deep learningdenoisingmachine learningmicrowavemodulatorreceiverrecurrent neural network (RNN)signal classificationsignal processingvariational autoencoder (VAE)

Autor:innen und Institutionen

  • Department of Mathematics and Industrial Engineering, Polytechnique Montreal, Montreal, QC H3T 1J4, Canada.
  • Institut National de la Recherche Scientifique (INRS), Energie Materiaux et Telecommunications, Montreal, QC H5A 1K6, Canada.

ORCID-Profile: Homa Arab, Serioja O Tatu, Steven Dufour

Zitieren

APA

Arab, H., Suzandeh, N., Tatu, S. O., & Dufour, S. (2026). A Variational Autoencoder Network for Heartbeat Signal Reconstruction and Breathing-State Classification Using 24 GHz Radar Sensors. Sensors, 26(21), 7083. https://doi.org/10.3390/s26217083

Vancouver

Arab H, Suzandeh N, Tatu SO, Dufour S. A Variational Autoencoder Network for Heartbeat Signal Reconstruction and Breathing-State Classification Using 24 GHz Radar Sensors. Sensors. 2026;26(21):7083. doi:10.3390/s26217083

BibTeX
@article{Arab2026,
  author = {Arab, Homa and Suzandeh, Nima and Tatu, Serioja O and Dufour, Steven},
  title = {A Variational Autoencoder Network for Heartbeat Signal Reconstruction and Breathing-State Classification Using 24 GHz Radar Sensors},
  journal = {Sensors},
  year = {2026},
  volume = {26},
  number = {21},
  pages = {7083},
  doi = {10.3390/s26217083}
}
Kostenlos starten →