An effective method for component activity classification supporting location awareness and user identification.

Area of Science:

• Human-Computer Interaction
• Wearable Technology
• Ubiquitous Computing

Background:

• Current activity recognition systems often lack location awareness and user identification.
• Distinguishing between similar activities (e.g., sitting on different furniture) remains a challenge.

Purpose of the Study:

• To develop an effective method for component activity classification that integrates location awareness and user identification.
• To improve the accuracy of recognizing user activities in a home environment.

Main Methods:

• A three-module system comprising a Pressure Sensing Module (PSM), Activity Detecting Module (ADM), and Receiving Station (RS).
• The wearable ADM classifies atomic activities (lie, sit, stand) and interactions with furniture.
• A short-range RF chip in the PSM (approx. 1m) enables proximity detection to furniture, facilitating location awareness.

Main Results:

• Achieved 92.5% average precision in component activity classification.
• Successfully integrated user identification, location awareness, and activity recognition.
• Demonstrated the system's ability to differentiate activities based on furniture interaction.

Conclusions:

• The proposed system effectively classifies component activities while simultaneously providing location awareness and user identification.
• This approach enhances the understanding of user context in smart environments.
• The high precision achieved suggests practical applicability in various user-centric systems.