EUVIMEDEUVIMED entdecken
Crossref

HEBR: An HRV Noncontact Health Monitoring Method Based on Millimeter-Wave Radar-Enhanced Time–Frequency Concentration Analysis

Jinsong Li, Xingguang Li, Yujian Cai, Kaiyao Shi

IEEE Internet of Things Journal · 15.04.2026 · 13 (8) · 17376-17392

DOI: 10.1109/jiot.2026.3661997 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Internet of Things Journal, erschienen am 15.04.2026. Verfasst von Jinsong Li und 3 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 42 Literaturangaben. Gefördert von: Science and Technology Development Plan Project of Jilin Province.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Noncontact heart rate variability (HRV) monitoring using frequency-modulated continuous-wave (FMCW) radar holds strong potential for bedside and home environments. Accurate interbeat interval (IBI) estimation is essential for reliable HRV assessment and early cardiovascular risk screening. However, under low signal-to-noise ratio (SNR) conditions with respiratory and motion interference, existing approaches are fundamentally limited by time–frequency uncertainty, making it difficult to stably recover high-precision IBI estimates, which restricts the practical deployment of noncontact HRV techniques. To enable high-precision IBI detection, this article proposes a harmonic-aware enhanced beat reconstruction (HEBR) method. HEBR constructs a time–frequency representation with enhanced time–frequency concentration to aggregate heartbeat energy under noise. It then employs dynamic programming to track the fundamental heartbeat component and form a harmonic support set. Phase-consistent band-limited reconstruction is performed on this support set to obtain a high-fidelity beat waveform. Using electrocardiogram (ECG) signals as the reference, evaluations on radar data from ten healthy subjects and one cardiac patient show that HEBR achieves median absolute errors (MedAEs) of 27.93, 15.37, and 7.24 ms for the IBI, the root mean square of successive differences (RMSSDs), and the standard deviation of normal-to-normal (NN) intervals (SDNNs), respectively. Additional experiments on nine hospitalized cardiac patients with photoplethysmogram (PPG) reference further suggest good cross-population generalizability. The beat recall, precision, and $F1$ -score relative to ECG are 88.4%, 90.1%, and 0.893, demonstrating robust performance under realistic conditions. HEBR provides a physically interpretable and readily deployable sensing solution for long-term, low-engagement HRV monitoring in Internet of Things (IoT) environments, and is promising for downstream applications such as bedside monitoring and emotion recognition.

Förderung

  • Science and Technology Development Plan Project of Jilin Province

Zitieren

APA

Li, J., Li, X., Cai, Y., & Shi, K. (2026). HEBR: An HRV Noncontact Health Monitoring Method Based on Millimeter-Wave Radar-Enhanced Time–Frequency Concentration Analysis. IEEE Internet of Things Journal, 13(8), 17376-17392. https://doi.org/10.1109/jiot.2026.3661997

Vancouver

Li J, Li X, Cai Y, Shi K. HEBR: An HRV Noncontact Health Monitoring Method Based on Millimeter-Wave Radar-Enhanced Time–Frequency Concentration Analysis. IEEE Internet of Things Journal. 2026;13(8):17376-17392. doi:10.1109/jiot.2026.3661997

BibTeX
@article{Li2026,
  author = {Li, Jinsong and Li, Xingguang and Cai, Yujian and Shi, Kaiyao},
  title = {HEBR: An HRV Noncontact Health Monitoring Method Based on Millimeter-Wave Radar-Enhanced Time–Frequency Concentration Analysis},
  journal = {IEEE Internet of Things Journal},
  year = {2026},
  volume = {13},
  number = {8},
  pages = {17376--17392},
  doi = {10.1109/jiot.2026.3661997}
}
Kostenlos starten →