Short Description

In the TU Graz LEAD project ‘Dependable Internet of Things’, a small-scale positioning system has been developed based on the low-cost DW1000 frontend (see picture). We call this system SALMA (single anchor localization with multipath assistance). SALMA is able to perform indoor positioning in a static environment with known floorplan, reaching decimeter level accuracy while only using a single anchor / base station.

The aim of this thesis is to extend the existing system by tracking methods (e.g., Kalman / particle filters), i.e. to basically give SALMA a memory.
SALMA currently only uses single snapshots of the channel impulse response as observations to estimate the position of the agent device. In mobile scenarios where the agent is moving around, it would be highly beneficial to take previous observations into account as prior knowledge for the positioning algorithm.

The full framework and system will be available for the student to go wild with the development of various extensions. Additionally, we performed an extensive measurement campaign in a parkhouse, and the collected data should provide a great basis to explore and test any algorithm that comes to mind.

Your Tasks

• Get familiar with the SALMA framework and the measured signals (channel impulse responses) that are analyzed in MATLAB (many detailed scripts already exist)
• Develop and implement additional methods in MATLAB that make use of the signals in different scenarios
• Examine and compare your findings with state-of-the-art methods

Your Profile

• Motivation and reliability
• Interest and knowledge in wireless communications and localization (and possibly basic knowledge in HF engineering)
• Basic knowledge in Matlab programming