3D Indoor Positioning in 5G networks

Over the past two decades, the challenge of accurately positioning objects or users indoors, especially in areas where Global Navigation Satellite Systems (GNSS) are not available, has been a significant focus for the research community. With the rise of 5G IoT networks, the quest for precise 3D positioning in various industries has driven researchers to explore various machine learning-based positioning techniques.
Within this context, researchers are leveraging a mix of existing and emerging wireless communication technologies such as cellular, Wi-Fi, Bluetooth, Zigbee, Visible Light Communication (VLC), etc., as well as integrating any available useful data to enhance the speed and accuracy of indoor positioning. Methods for indoor positioning involve combining various parameters such as received signal strength (RSS), time of flight (TOF), time of arrival (TOA), time difference of arrival (TDOA), direction of arrival (DOA) and more.
Among these, fingerprint-based positioning stands out as a popular technique in Real Time Localisation Systems (RTLS) due to its simplicity and cost-effectiveness.
Positioning systems based on fingerprint maps or other relevant methods find applications in diverse scenarios, including malls for indoor navigation and geo-marketing, hospitals for monitoring patients, doctors, and critical equipment, logistics for asset tracking and optimising storage spaces, and homes for providing Ambient Assisted Living (AAL) services.
A significant challenge facing all indoor positioning systems is the objective evaluation of their performance. This challenge is compounded by the coexistence of heterogeneous technologies and the rapid advancement of computation. There is a vast potential for information fusion to be explored. These observations have led to the motivation behind our work. As a result, two novel algorithms and a framework are introduced in this thesis.