Monitoring Respiratory Motion with Wi-Fi CSI: Characterizing performance and the BreathSmart Algorithm

Published

December 15, 2022

Author(s)

Susanna Mosleh, Jason Coder, Keith Forsyth, Mohamad Omar Al Kalaa, Christopher Scully

Abstract

Objective: Respiratory motion (i.e., motion pattern and rate) can provide valuable information in many medical situations. This information may help diagnose different health disorders and diseases. Wi-Fi-based respiratory monitoring schemes utilizing commercial off-the-shelf (COTS) devices can provide contactless, low-cost, simple, and scalable respiratory monitoring without requiring any specialized hardware. Despite intense research efforts, an in-depth investigation of how to evaluate this type of technology is missing. Methods: In this paper, we demonstrate and assess the feasibility of monitoring and extracting human respiratory motions and rates from Wi-Fi channel state information (CSI) data. This demonstration consists of the implementation of an end-to-end system of a COTS-based hardware platform, control software, data acquisition, and a proposed processing algorithm. The processing algorithm is a novel deep learning-based approach that exploits both CSI amplitude and phase information to reveal unique characteristics of different minute movements. These small changes in CSI values can be processed to learn high-level abstractions of breathing-induced chest movements–leading to an accurate respiratory pattern and rate estimation. We also conduct extensive laboratory experiments and measurements that demonstrate an evaluation technique that could be replicated when quantifying the performance of similar systems. Results: Results indicate that our proposed scheme can classify respiratory patterns and rates with an accuracy of 99.54% and 98.69%, respectively, in moderately degraded RF channels. We also examine the impact of RF path-loss and CSI frame rate on classification accuracy. Significance: Understanding the feasible limits and potential failure factors of the Wi-Fi CSI-based respiratory monitoring scheme — and how to evaluate them — is an important step toward the practical deployment of this technology. This study discusses ideas for further expanding this technology.

Citation

IEEE Transactions on Biomedical Engineering

Pub Type

Journals

Download Paper

Local Download

Keywords

Channel state information, deep learning, MIMO-OFDM, LSTM, respiration monitoring, respiratory motion classification, Wi-Fi.

Citation

Mosleh, S. , Coder, J. , Forsyth, K. , Al Kalaa, M. and Scully, C. (2022), Monitoring Respiratory Motion with Wi-Fi CSI: Characterizing performance and the BreathSmart Algorithm, IEEE Transactions on Biomedical Engineering, [online], https://tsapps.nist.gov/publication/get_pdf.cfm?pub_id=935107 (Accessed December 30, 2024)