Journal article icon

Journal article

VitalCSI: Contactless Respiratory Rate Estimation Using Consumer-Grade Wi-Fi Channel State Information

Abstract:
Continuous respiratory rate (RR) monitoring can improve the detection of clinical events, such as pulmonary infections, cardiac arrests, and sleep apnoea. Wi-Fi-based systems offer a low-cost, contactless alternative to radar and video. However, existing studies are limited to narrow respiratory ranges and small-scale validation. We present VitalCSI, a vital sign monitoring system using off-the-shelf, low-power Wi-Fi hardware. We recorded 15 healthy university athlete volunteers and developed RR estimation algorithms benchmarked against nasal airflow sensors. VitalCSI uses a consumer Wi-Fi access point and a Raspberry Pi computer to capture channel state information (CSI). We estimated the RR from CSI via principal component analysis (PCA), spectral peak detection, and breath (counting in 30 s windows), which were then fused by a multidimensional Kalman filter. VitalCSI showed strong agreement with airflow references (r2=0.93, MAE = 1.20 brpm), tracking RR across 6–33 brpm and outperforming prior Wi-Fi studies. VitalCSI demonstrates the feasibility of RR monitoring with a single-antenna, single-board microcomputer as the Wi-Fi transmitter. It is the first validated system for continuous, contactless RR monitoring using consumer-grade Wi-Fi over an extended respiratory range, paving the way for use in both home and sports monitoring contexts.
Publication status:
Published
Peer review status:
Peer reviewed

Actions

Access Document

Publisher copy:
10.3390/s26010225

Authors

More by this author
Institution:
University of Oxford
Division:
MPLS
Department:
Engineering Science
Sub department:
Institute of Biomedical Engineering
Role:
Author
More by this author
Institution:
University of Oxford
Division:
MPLS
Department:
Engineering Science
Sub department:
Institute of Biomedical Engineering
Role:
Author
ORCID:
0000-0003-3723-8103
More by this author
Institution:
University of Oxford
Role:
Author
ORCID:
0000-0001-9862-4732
More by this author
Institution:
University of Oxford
Role:
Author
ORCID:
0000-0003-4787-6053


Publisher:
MDPI
Journal:
Sensors More from this journal
Volume:
26
Issue:
1
Pages:
225
Article number:
225
Publication date:
2025-12-29
Acceptance date:
2025-12-24
DOI:
EISSN:
1424-8220
ISSN:
1424-8220


Language:
English
Keywords:
Pubs id:
2356806
UUID:
uuid_4935e9bb-33d0-41e8-af97-eebb63e247b0
Local pid:
pubs:2356806
Source identifiers:
3643648
Deposit date:
2026-01-08
ARK identifier:
This ORA record was generated from metadata provided by an external service. It has not been edited by the ORA Team.

Terms of use


Views and Downloads






If you are the owner of this record, you can report an update to it here: Report update to this record

TO TOP