Bluetooth-based indoor positioning systems | |
| Author | Apiruk Puckdeevongs |
| Call Number | AIT Diss. no.RS-20-04 |
| Subject(s) | Indoor positioning systems (Wireless localization) Bluetooth technology |
| Note | A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Remote Sensing and Geographic Information Systems |
| Publisher | Asian Institute of Technology |
| Abstract | In recent days, Indoor Positioning System (IPS) has become famous and widely used technique in locating position problem. IPS has applied to various types of wireless communications technology such as Wireless Local Area Network (WLAN), Zig bee, and ultrasound. In IPS research space, Bluetooth is one of the leading considered technologies because of the following key benefit Bluetooth is commonly available in the market. First, it is low price compared to other similar technologies. Second, it is quick and quite easy to setup and deploys. Third, Bluetooth can last longer with small power consumption, and finally it can integrate with mobile device such as personal smartphones. These essential benefits demonstrate Bluetooth technology can be instrumental with low hardware and setup cost. For Bluetooth indoor positioning, the range estimation of the positioning by the Bluetooth Receive Signal Strange Indicator (RSSI) use in this study .. However, the RSSI values varied within the building and each measured value is unstable, the fingerprinting technique should be used to create a database of signals and calculate the coordinates using Artificial Neural Network (ANN). Due to the complexity of RSSI measurements in the building, it is difficult to write in terms of mathematical relationships. This study applies Feed-Forward Multilayer Perceptron (FF -MLP) as the primary model for position estimation. The experiment to find an appropriate FF -MLP structure for position estimation problem in the experimental area. has the following conditions the structure must be small and provide sufficient accuracy. Apart from that, , this study continues to improve accuracy by considering using Radial basis function (RBF) network to work with FF-MLP. The result from applying RBF network to the FF -MLP increases in accuracy by 13.28% compared to using FF -MLP stand alone which means that the model using FF-MLP and RBF give the result with less than 1 metre deviation at 70% . Moreover, this research uses FF -MLP and RBF model to also apply with another area of study which yields the similar result. Therefore, this additional study assures that the proposed model performs well in positioning system. |
| Year | 2020 |
| Type | Dissertation |
| School | School of Engineering and Technology (SET) |
| Department | Department of Information and Communications Technologies (DICT) |
| Academic Program/FoS | Remote Sensing (RS) |
| Chairperson(s) | Tripathi, Nitin Kumar |
| Examination Committee(s) | Apichon Witayangkurn;Poompat Saengudomlert |
| Scholarship Donor(s) | Royal Thai Government Fellowship |
| Degree | Thesis (Ph.D.) - Asian Institute of Technology, 2020 |