Year: 2026
Type: article
Abstract: Contactless vital sign monitoring using millimeter-wave (mmWave) radar holds immense promise for ubiquitous healthcare. However, contemporary research is heavily constrained by what we define as the Static Trap-an idealized experimental paradigm where subjects must remain completely motionless during evaluation. This oversimplification overlooks continuous body movement (CBM) that are inevitable in real-world clinical and home-care settings. From a signal-processing perspective, CBM introduces high-amplitude non-linear motion artifacts, triggering catastrophic phase unwrapping failures and severe spectral smearing that render traditional linear filtering pipelines entirely ineffective. This article address this persistent and unresolved bottleneck. Drawing on recent algorithmic insights into spatial multi-point redundancy and blind source separation (specifically non-negative matrix factorization), we discuss the feasibility of decoupling weak, quasi-periodic physiological micro-motions from high-amplitude macroscopic motion artifacts. Finally, we call upon the research community to shift its evaluation metrics from static marginal gains to motion-robustness boundaries, develop standardized open-source CBM datasets, and fuse radar signal processing with kinematics priors to transition contactless monitoring from laboratory novelties to dependable clinical tools.
Source: Frontiers in Signal Processing
Authors: Tianyi Wang(Conceptualization, Data curation, Formal analysis, Funding acquisition, Methodology, Project administration, Visualization, Writing – original draft), Wenxu Sun(Data curation, Methodology, Project administration, Writing – original draft), Ryo Shirai(Methodology, Validation, Writing – original draft), Keisuke Shima(Funding acquisition, Writing – original draft)
Institutions: Yokohama National University, Kyoto University
Language: English
Subfield: Biomedical Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: gold
Estimated APC paid: $2,353
OpenAlex ID: W7233997751
