Design And Implementation Of Wearable Devices Used For Elderly Monitoring

Today more elderly tends to live alone.Unfortunately they are suffering more accidents,which brings significant threat to their health or even their lives.Therefore,monitoring of elderly people living alone now attracts wide attention of the society,especially of their children.Guardians are always more relax when equipped with the monitor out of trust.As a result,reliability is particularly important for the elderly monitoring system,for any miss,even a little one,it may causes disastrous consequences.Currently special-monitoring equipment is designed to achieve the primary monitoring function,but tends to ignore the device size,battery life,user experience,and other aspects.Those smart bracelet,smart watches,smart phone products that equipped with different kinds of sensors are designed for fitness,entertainment,or communications,but not the elderly people's monitoring.Obviously they cannot meet the requirements when used as monitors.Considering the urgent needs of society,it is very necessary to design an ergonomic,compact and wearable monitoring system to guard the elderly people's safety.This paper analyzes the advantages and shortcomings of existing monitoring systems,carries out researches in many aspects,such as overall program design,device selection,design,hardware design,and software design,to improve the reliability.Aims at the elderly monitoring and takes reliable as the principle,this paper designs a lasting wearable system and implements the core of the wearable device.The main contents are as follows:(1)This paper designs and produces wearable elderly monitoring device using 3D printing technology.The device' s 4-layer PCB board is integrated with microprocessor,a Bluetooth module and accelerometer,gyroscope,magnetometer,barometer and some other devices.The overall size of the device is 8mm X 27mm X 27mm,which significantly improves users' wearing experience.(2)Based on the hardware platform,this paper proposes online real-time solver of body posture.According to the calculated data,the paper conducts a recognition and classification of user's daily behavior by using pattern recognition method.(3)By analyzing the changing and statistical characteristics of users' abnormal behavior,the paper proposes fall detection algorithm based on multi-sensor data fusion,so that carries out fall detection unnecessary for fixed positions.(4)This paper programs a software based on Android system which has many functions such as computer interaction,Bluetooth communication,positioning,abnormal behavior alarm,exercise analysis.