Study, implementation and measurements of bluetooth low energy indoor positioning system with client/eerver infrastructure
Indoor positioning is an important research topic today, and many different technologies are investigated in numerous projects as well as standardization bodies. Some of these technologies require a special map. These maps may contain fingerprints of radio-frequency (RF) signals such as Bluetooth or...
Saved in:
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Language: | en_US |
| Published: |
2017
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/11610/17741 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Indoor positioning is an important research topic today, and many different technologies are investigated in numerous projects as well as standardization bodies. Some of these technologies require a special map. These maps may contain fingerprints of radio-frequency (RF) signals such as Bluetooth or Wi-Fi (often called radio maps) or unique characteristics of the magnetic field. The collection of data by many users (crowdsourcing) to create or update such maps is a promising approach to reduce the effort and cost of the map creation. In this master thesis, a server-client architecture that collects sensor data (inertial sensors, RF signal strengths, etc.) from smartphone users was studied and developed in order to enable the development of crowdsourcing algorithms and architectures. The sensor data have been collected with smartphone devices with an android client, named InLoCS, and the sensor data were transmitted wirelessly to a MySQL database via a REST HTTP API programmed in NodeJS. The server is capable of storing and retrieving the collected data with JSON objects and a suitable relational database format is maintained to correlate the data with appropriate metadata information that describes each user, session and sensor. Furthermore, a Bluetooth Low Energy position estimation algorithm was created and evaluated by using the pre-recorded data. This positioning algorithm was intended to be a part of the building block of a fusion Indoor Localization algorithm developed by Sony and is able to estimate the position of the user with a mean localization error of 2.15 meters. |
|---|