This paper proposes an android-based application to help the users to navigate in finding books in the library easily and interactively. This navigation application is connected to a Bluetooth Low Energy (BLE) device that will emit an RSSI signal received by the Smartphone user and show the desired distance to the bookshelf position. The method of triangulation and mean filter were used to eliminate noise in the test environment to make the position of the bookshelf can be found precisely based on the RSSI BLE Beacon value. The test results showed the largest RSSI value for LOS conditions at -48dBm and NLOS at -63 dBm; while the lowest RRSI values for LOS conditions was at -84dBm and NLOS was at -96dBm.

The key to fingerprint positioning algorithm is establishing effective fingerprint information database based on different reference nodes of received signal strength indicator (RSSI). Traditional method is to set the location area calibration multiple information sampling points, and collection of a large number sample data what is very time consuming. With Zigbee sensor networks as platform, considering the influence of positioning signal interference, we proposed an improved algorithm of getting virtual database based on polynomial interpolation, while the pre-estimated result was disposed by particle filter. Experimental result shows that this method can generate a quick, simple fine-grained localization information database, and improve the positioning accuracy at the same time.