Design And Realization Of Human Movement Monitoring System Based On Body Sensor Network

Nowadays there are quite a number of sick, injured and disabled people who need to receive rehabilitation therapy by exercise training. Quantitative monitor of movement parameters and movement intervention are important for their rehabilitation therapy. However, currently a lot of physical disabled people are unable to receive rehabilitation therapy because of shortage of rehabilitation physiatrists and expensive rehabilitation therapy fee in our country. With changes of social structure, all the countries in the world have to face the problem of aging population. Because of gradual degeneration of physiological functions, movement abilities of old people decrease and the probability of falling becomes high. Besides death, falling also brings much disability and affects physically and mentally healthy life for old people. With aging population aggravating, if without effective measures, medical burden caused by old people falling will become heavier and heavier.As a branch of wireless sensor network and an important component of internet of things, body sensor network (BSN) incorporates biomedical engineering technology and wireless sensor network and has the characteristics of high precision, narrow monitor range, miniaturization, low power consumption, easy to operate, safety and so on. It can monitor human physiological information and movement information real-timely and provide a lot of vital quantitative information to testers and then achieve earlier monitoring, earlier prevention and earlier intervention for some common chronic diseases and can be widely used in the fields of biomedical engineering, athletic training, daily movement monitoring and so on. Besides, BSN is also a realization form of home health care and provides a brand-new thought for telemedicine and home health care.Based on above research background, this thesis aimed at realizing a human movement monitoring system based on body sensor network and took low cost, low power consumption, high modularization, high precision, convenient to wear and so on as guideline. A human movement monitoring system based on nine-axis wireless sensor platform was developed and the object of real-time movement monitoring was initially realized. This thesis is organized as follows:Firstly, a human movement monitoring system based on nine-axis wireless sensor platform was constructed and the system included hardware, embedded software and upper machine software. The system contained3-axis accelerometer,3-axis magnetometer,3-axis gyroscope, nRF905wireless transceiver, MSP430F2418microcontroller and so on and provided9-DOF movement data real-timely and continuously.Secondly, human regular walking experiment was carried out and walking acceleration signal was taken as an example to compare four common filters for movement data:median filter, Butterworth low-pass filter, discrete wavelet package shrinkage and kalman filter. Output of motion tracker MTx from Xsens was regarded as reference signal. A preliminary result showed that kalman filter had good performance in real-time, waveform delay and filtering noise and was the most suitable for filtering acceleration signal for the platform under the condition of low velocity movement.Thirdly, combination of acceleration and magnetic signal and gyroscope signal integral were adopted respectively to calculate attitude angle. Two simple experiments were carried out to analyze calculation errors of attitude angle. Human skeleton movement and walking were attempted with real-timely calculated attitude angle and the object of real-time movement monitoring was initially realized. While attempting walking, double pendulum model which calculated component variation in adjacent time in walking direction of lower extremity was used to calculate displacement of linear walking.Finally, we made a conclusion and proposed the future research work for human movement monitoring system.