Design and Evaluation of a Solo-Resident Smart Home Testbed for Mobility Pattern Monitoring and Behavioural...

Area of Science:

• Gerontology
• Computer Science
• Ubiquitous Computing

Background:

• The growing elderly population requires solutions for independent living.
• Smart home technology offers unobtrusive monitoring of health and mobility patterns.
• Assessing elderly residents' well-being through activity data is crucial.

Purpose of the Study:

• To evaluate a smart home system testbed for monitoring solo-resident elderly.
• To optimize data collection for accurate mobility pattern analysis.
• To assess resident behavior using activity level data.

Main Methods:

• Deployment of paired Passive Infra-red (PIR) sensors at house entries.
• Utilizing process mining techniques, specifically the parallel activity log inference algorithm (PALIA), for post-deployment analysis.
• Conducting a re-deployment phase after analysis to improve data accuracy.

Main Results:

• Data accuracy significantly improved from 81.57% to 95.53% after re-deployment.
• The study demonstrated the usability of smart home data for behavioral analysis.
• Collected results showed similar patterns when compared to the CASAS project dataset.

Conclusions:

• Optimized smart home monitoring systems can effectively support elderly independence.
• Process mining techniques like PALIA are valuable for refining data collection in smart homes.
• Accurate mobility pattern data facilitates reliable behavior assessment for elderly care.