Wireless hazardous fall detection and location tracking for medical urgency and residence care applications

In this thesis, a wireless system for monitoring and identifying people who need special care and medical attention such as elders disabled persons and patents. Fall is one of the most dangerous accidents for these people, which may cause serious physical injury. The wireless system presented in this thesis can detect if a person needs help. A wearable fall detection sensor is presented and demonstrated in this work. When a serious fall is detected, the wireless network that covers the whole building can position that person and send this information to a central server.;A fall detection system employing triaxial accelerometers is analyzed, designed, fabricated and tested. This fall detection system makes use of a simple method mainly based on energy expenditure. This system does not require high performance DSP circuits. A simple microcontroller can be used for data processing. Measurements show this system works very well.;A wireless sensor network used for target positioning and wireless communication. The positioning may be performed based on received signal strength(RSS). This kind of indoor positioning systems usually are not as accurate as positioning systems based on time measurements. However, they are less expensive and are able to provide an acceptable accuracy for this work. Many works on positioning with RSS focus on current devices used in wireless local area networks. If there are no such wireless networks, people have to build such an IEEE802.11 network, which is specially designed for wireless communication other than for positioning. And those network cards are relatively expensive. As a result, for indoor environment where there is no need for such networks, an affordable wireless sensor network is specially designed for indoor positioning. Factors that affect the performance have been presented and analyzed. The performance of the whole system is analyzed based on measurements.;There are four chapters in this thesis. Chapter one gives the background of current fall detection methods and indoor positioning systems. Chapter two explains our fall detection system, including hardware design and the fall detection algorithm. Measurement results for this system are also presented. In chapter three, the architecture of our wireless sensor network will be introduced, including the protocols and the wireless unit. Chapter four talks about the positioning method and performance. Several parameters that can affect the positioning performance are analyzed. Results based on measurements and Monte-Carlo method are also shown there.