Goal of this thesis is to create an online service, suitable for approximation of user's position inside a building via his mobile device. For this goal a known network of Wi-Fi access-points was used in combination with a machine learning algorithms. Algorithms were trained and tested on the ground floor in building Q of Mendel University in Brno. The Algorithms and the online service were tested with automated tests as well as in real environment with a mobile application, which was also created as a part of this thesis. Results have shown that reliability and accuracy of service is highly dependent on localized mobile device. Success rate floated between 43~\% and 66~\%, depending on a mobile device.
| Identifer | oai:union.ndltd.org:nusl.cz/oai:invenio.nusl.cz:431512 |
| Date | January 2017 |
| Creators | Kalman, Ondřej |
| Source Sets | Czech ETDs |
| Language | Czech |
| Detected Language | English |
| Type | info:eu-repo/semantics/masterThesis |
| Rights | info:eu-repo/semantics/restrictedAccess |
Page generated in 0.0016 seconds