Implementation of accelerometer sensor module and fall detection monitoring system based on wireless sensor network

Area of Science:

• Biomedical Engineering
• Wearable Technology
• Sensor Networks

Background:

• Accurate monitoring of human posture and activity is crucial for healthcare and sports.
• Existing methods may lack real-time monitoring capabilities or user mobility.
• Fall detection systems are vital for elderly care and patient safety.

Purpose of the Study:

• To implement a wireless accelerometer sensor module and algorithm for determining wearer's posture, activity, and falls.
• To assess the performance and detection rates of the implemented system.
• To develop a real-time monitoring system for postures, motions, and falls.

Main Methods:

• Utilized an ADXL202, 2-axis accelerometer sensor module with a wireless RF module.
• Developed an algorithm analyzing AC and DC components of accelerometer signals for activity and posture recognition.
• Conducted experiments with 30 subjects to evaluate algorithm performance and detection rates.
• Implemented a wireless sensor network for monitoring in an experimental space.

Main Results:

• The ADL algorithm successfully differentiated between various postures (standing, sitting, lying) and activities (walking, running).
• Experimental assessments with 30 subjects yielded calculated detection rates for postures, motions, and individuals.
• Simulation experiments confirmed the system's capability in fall detection across multiple activities and trials.

Conclusions:

• The developed system is applicable for activity monitoring and fall detection in patients and the elderly.
• It offers potential for sports athletes' exercise measurement, pattern analysis, and general user fitness training.
• The system can serve as an entertainment tool for monitoring personal activities and play.