Scholarlink

<p>With the development of an aging population, frequency-modulated continuous-wave (FMCW) radar has gained traction in vital signs research due to its noncontact nature, particularly in indoor nursing environments. However, effective noncontact vital signs monitoring requires the automatic localization of human targets first, given the randomness of target position distribution and the fact that it cannot be known in advance. This study proposes an FMCW noncontact vital signs monitoring method that integrates accurate target localization with vital signs extraction, consisting of three functional modules. The target localization module employs a cell variance average of Constant False Alarm Rate (CVA-CFAR) and a Density-based Mean Clustering Algorithm (DMCA) to cluster multi-frame signals and accurately detect human targets. The phase signal preprocessing module uses a hybrid curve interpolation technique to address missing values in phase data, thereby enhancing signal integrity. In the vital signs extraction module, the Extended Kalman Filter (EKF) is utilized to accurately track respiratory rate (RR) and heart rate (HR). Comprehensive experiments were conducted with single and multiple human targets monitored by FMCW radar at varying distances. Results demonstrate that this approach enables precise automatic positioning, leading to more accurate and robust vital sign monitoring across diverse scenarios.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

FMCW Noncontact Vital Signs Monitoring with Accurate Positioning of Target Clustering and Interpolation

  • 期刊论文

  • 发表日期：

    2025年9月26日

查看全文
  • Mingjian Zhao,
  • Senlong Chen,
  • Jun Liao,
  • Laiqing Li

摘要

With the development of an aging population, frequency-modulated continuous-wave (FMCW) radar has gained traction in vital signs research due to its noncontact nature, particularly in indoor nursing environments. However, effective noncontact vital signs monitoring requires the automatic localization of human targets first, given the randomness of target position distribution and the fact that it cannot be known in advance. This study proposes an FMCW noncontact vital signs monitoring method that integrates accurate target localization with vital signs extraction, consisting of three functional modules. The target localization module employs a cell variance average of Constant False Alarm Rate (CVA-CFAR) and a Density-based Mean Clustering Algorithm (DMCA) to cluster multi-frame signals and accurately detect human targets. The phase signal preprocessing module uses a hybrid curve interpolation technique to address missing values in phase data, thereby enhancing signal integrity. In the vital signs extraction module, the Extended Kalman Filter (EKF) is utilized to accurately track respiratory rate (RR) and heart rate (HR). Comprehensive experiments were conducted with single and multiple human targets monitored by FMCW radar at varying distances. Results demonstrate that this approach enables precise automatic positioning, leading to more accurate and robust vital sign monitoring across diverse scenarios.