Abstract
This study presents a non-contact continuous blood pressure (BP) estimation method utilizing a customized K-band continuous-wave radar sensor. Four time-domain features, extracted from the radar-detected wrist pulse, serve as key parameters for BP estimation. Additionally, an adaptive filtering technique is implemented to effectively mitigate low-frequency noise. A support vector machine technique is employed for BP estimation. The experimental protocol involved collecting continuous daytime records from six participants over three days. Results demonstrate the efficacy of the proposed adaptive filtering BP estimation method. These findings underscore the potential of the proposed method for future reliable non-contact BP monitoring applications.