Non-Contact Smart Sensing of Physical Activities during Quarantine Period Using SDR Technology.

Area of Science:

• Engineering
• Computer Science
• Health Technology

Background:

• The COVID-19 pandemic necessitates continuous monitoring of physical activity and fall detection, especially for individuals with health issues.
• Wearable sensors are impractical during pandemics, highlighting the need for non-contact monitoring solutions.
• Physical activity monitoring is crucial for managing health conditions exacerbated by sedentary lifestyles during quarantine.

Purpose of the Study:

• To develop and evaluate a non-contact smart sensing platform for monitoring human physical activities and detecting falls during quarantine.
• To leverage software-defined radio (SDR) technology for intelligent, flexible, and portable health monitoring.
• To explore the potential of wireless channel state information (WCSI) for activity classification.

Main Methods:

• Developed a non-contact smart sensing through the walls (TTW) platform utilizing software-defined radio (SDR).
• Exploited orthogonal frequency division multiplexing (OFDM) signals with 64-subcarrier wireless channel state information (WCSI).
• Applied machine learning algorithms, specifically a fine tree algorithm, for activity classification.

Main Results:

• The TTW platform demonstrated intelligent, flexible, portable, and multi-functional capabilities.
• Achieved a 99.7% accuracy in classifying fall activity separately from standing, walking, running, and bending.
• The system effectively utilizes WCSI for distinguishing various physical activities.

Conclusions:

• Non-contact smart sensing using SDR technology is a viable and highly accurate method for monitoring physical activities and detecting falls.
• This technology offers a promising solution for continuous health monitoring during quarantine and pandemic situations.
• The developed platform opens new research avenues for detecting COVID-19 symptoms and monitoring various diseases remotely.