Author:

Abstract:

In current healthcare environments, such as hospitals and other skilled nursing facilities, a trend towards mobile and personalized interactions between patients and medical staff has recently emerged. Personalized care is capable of increasing the patient comfort feeling, since he or she is adequately assisted within a limited amount of time, while also the care giving staff benefits from location based services: automated access control, automated availability of the patient's care files, assistance in drug administering, etc. Even though these personalized, location based services are clearly time preserving for the care giving staff, technological constraints combined with economical considerations cause these services to still be very little implemented in the current care facilities. The most considerable technological barrier is the fact all people present in the care facility (nurses, physicians, but also patients and technology assisting staff) should be located throughout the care facility within a reasonable amount of time, or even more challenging: continuously tracked. The present doctoral research focuses on the development of an indoor localization system specifically designed for use in healthcare facilities, which are characterized by long corridors and small, more or less equally sized rooms at each side of this corridor. Therefore, the field of application hence includes hospitals with possibly thousands of patient beds, but also smaller topologies as can be found in rest homes. The developed system must by all means by scalable, and provide accurate localization up to room level (or locate people in a certain sector of a corridor). The research focuses on a method by which a mobile unit worn by a person is capable of locating itself indoors, based on signal strength measurements which can be related to RF beacons installed at predetermined positions in the care facilities. These predetermined positions correspond to the actual location of ubiquitously available nurse call network access points, thus ensuring easy integration and cost effectiveness. For reasons of location update speed, only a minimal amount of simple yes/no decisions is needed. No prerequisites regarding the used hardware are required, except for an RF chip capable of performing RSS measurements. With extensive testing and statistical support, we prove the algorithm can be used in a healthcare setting with an envisioned level of localization accuracy up to room level (or region level in a corridor), while avoiding heavy investments. To simulate the performance of the developed indoor localization algorithm, a method for simulating signal strength measurements is developed, in which several building related parameters (room sizes, etc.) and RF related parameters (path loss exponents, EM wave attenuation due to wall penetration, etc.) can be entered. Hence, a performance prediction tool is available and can be used for other RF chips than the one used in this research. The mobile device is able to reason on information captured from beacons in the used communication channel, on condition that the sent information contains valid information about the beacon's position. An automated method for ensuring the information is correct, is described in this work: an automatic commissioning system of newly deployed beacons is proposed. Newly deployed beacons gather signal strength measurements between themselves while the commissioning system uses this centrally collected information to assign a secondary IEEE MAC48 address to each RF beacon in an organized way.