An Ultrasonic and Air Pressure Sensing System for Detection of Behavior before Getting out of Bed Aided by Fuzzy Theory

Hayato YAMAGUCHI  Hiroshi NAKAJIMA  Kazuhiko TANIGUCHI  Syoji KOBASHI  Yutaka HATA  

Publication
IEICE TRANSACTIONS on Information and Systems   Vol.E93-D   No.3   pp.542-549
Publication Date: 2010/03/01
Online ISSN: 1745-1361
DOI: 10.1587/transinf.E93.D.542
Print ISSN: 0916-8532
Type of Manuscript: PAPER
Category: Computation and Computational Models
Keyword: 
health monitoring,  ultrasonic oscillosensor,  air pressure sensor,  behavior monitoring system,  fuzzy inference,  

Full Text: PDF(1MB)>>
Buy this Article




Summary: 
This paper proposes a sensing system for a behavior detection system using an ultrasonic oscillosensor and an air pressure sensor. The ultrasonic oscillosensor sensor has a cylindrical tank filled with water. It detects the vibration of the target object from the signal reflected from the water surface. This sensor can detect a biological vibration by setting to the bottom bed frame. The air pressure sensor consists of a polypropylene sheet and an air pressure sensor, and detects the pressure information by setting under the bed's mattress. An increase (decrease) in the load placed on the bed is detected by the increase (decrease) in the pressure of the air held in the tube attached to the sheet. We propose a behavior detection system using both sensors, complementally. The system recognizes three states (nobody in bed, keeping quiet in bed, moving in bed) using both sensors, and we detect the behavior before getting out of bed by recognized these states. Fuzzy logic plays a primary role in the system. As the fundamental experiment, we applied the system to five healthy volunteers, the system successfully recognized three states, and detected the behavior before getting out of bed. As the clinical experiment, we applied the system to four elderly patients with dementia, the system exactly detected the behavior before getting out of the bed with enough time for medical care support.