| dc.description.abstract | Wireless Sensor Network (WSN) is one of the popular research subjects in recent years, especially for the indoor positioning application. Many solutions are applied in the following, for instances, Global Positioning System (GPS), Angle-Of-Arrival (AOA), Time-Of-Arrival (TOA), Time-Difference-Of-Arrival (TDOA) and Received Signal Strength (RSS), etc. Because of signal reflection, wall attenuation and extra hardware required, frontal four methods are suitable for outdoors only. On the contrary, RSS is an instinct approach and attractive for indoor environment.
In the past years, the application of location estimation focuses on locating an observed target by given several reference landmarks. However, we bring up a fascinating idea of using wireless sensor nodes to automatically construct the floor plan of an unknown indoor space. To our best knowledge, this is the first investigation to this field. Once floor plan is constructed and displayable, we can then precisely know the location of deployed sensor nodes and control/monitor any sensor node,
such as temperature/humidity reporting or on-off switching, by selecting/observing the location GUI window.
In the thesis, we first measure the characteristics of RSS as a function of the distance. After then, we introduce the room construction algorithm. As the floor plan construction is done, the methods of location positioning are addressed. For evaluations, we use MWNL (Mobile and Wireless Network Lab) Lab in NCU (National Central University) as an experimental example to show the derived results. | en_US |