Acoustic fall detection using a circular microphone array.

Area of Science:

• Gerontology
• Acoustic Engineering
• Signal Processing

Background:

• Falls are a significant health concern for individuals aged 65 and older, affecting over one-third annually in the US.
• Existing acoustic fall detection systems face challenges with false alarms in noisy and reverberant settings.

Purpose of the Study:

• To develop and evaluate an improved acoustic fall detection system (FADE) for elderly individuals.
• To enhance the accuracy and robustness of fall detection by utilizing an advanced microphone array configuration.

Main Methods:

• Development of the FADE system incorporating an 8-microphone circular array.
• Comparison with a previous 3-microphone linear array system to mitigate false alarms from sounds above floor level.
• Testing in noisy and reverberant environments to assess 3-D sound localization capabilities.

Main Results:

• The 8-microphone circular array demonstrated reliable and robust sound location estimation, even with interference.
• Preliminary experiments yielded encouraging classification results on a pilot dataset comprising 55 falls and 120 non-fall sounds.

Conclusions:

• The enhanced FADE system with a circular microphone array shows promise for accurate and reliable acoustic fall detection in challenging environments.
• Further development and validation are warranted to confirm its efficacy in real-world scenarios for elderly fall prevention.