Internet of things (IoT) is the network of devices which expected to be the dominant technology used by operators in these coming years due to its promising solutions for healthcare and transportation. The idea of the indoor positioning system (IPS) comes from the global positioning system (GPS) where people want to know the target position or want to track car movements. Unfortunately, GPS is not working in a closed building where direct line-of-sight (LOS) is required between the tracked device and the satellites. As a solution for that, different companies decided to build up a new tracked system which can be used for indoor. Nowadays, IPS is required in many closed places such as factories, hospitals, and some laboratories. Sunstone real-time localization system (RTLS) is a sophisticated IPS which made by OMTLAB Ltd and it has been installed in different factories in Hungary. Besides that, RTLS has some limitations in identifying the height of the tracked tag.
In this project, optimization of RTLS is applied using MEMS pressure sensor. Sunstone RTLS is based on time difference of arrival (TDOA) measurements, and with typical installations, it does not give a good estimation for the tag’s height due to this TDOA geometry problem. By taking the benefits of the relationships between the air pressure and altitude, MEMS pressure sensor is used as a solution which gives more accuracy compared to the old system. DPS310 miniaturized digital barometric air pressure sensor from Infineon technologies is used in this project which actually suitable for indoor navigation application because of it has low power consumption, high accuracy level and it has also a temperature sensor. Different scenarios of tests were applied in this project to measure the performance and efficiency of the pressure sensor with tracked tag. The scenarios of tests consist of position tests, day tests, pressure shock tests (Door open-close test), temperate tests, offset calibration and also calculate the height of the tag. The project is also included some statistical error analysis and then concluded by the proposed IPS system with the pressure sensor.