Location Prediction in 4G Systems

OData support
Dr. Imre Sándor
Department of Networked Systems and Services

The first step in my thesis work was research. I had familiarize myself with the history of positioning, all the methods that had been used before and I had to investigate the opportunities that new methods can offer. Furthermore I had to study the capabilities of LTE, the positioning methods that are in use nowdays in LTE.

A simulation framework was implemented in MATLAB environment. This framework was made in cooperation with one of me collegues. The main guidelines of this framework were modular structure, ease of development, it should meet the 4G mobile networks standards and it should support the implementation of as many methods as possible. The system’s fundamental building block was a channel model, the accuracy of this model had an impact on the accuracy of the whole work. So I started reasearch into the field of MIMO channel models and I acquired detailed knowledge in the field. For a while I thought about developing my own channel model, but this proved to be a huge task and even if I had succeeded there wouldn’t have been any guarantee for success in accuracy. So I continued on with searching for a usable channel model. There were some promising models, but they proved to be unsuitable for my purposes, finally I found the WINNER Phase II model that met my requirements. Thereupon I concentrated on the time-based methods, since I saw the greatest opportunities in them. I implemented two positioning algorithms, with which I managed to get accuracy below 50 meters during my measurements. There were some weaknesses to the algorithms that I studied and tried to overcome, with one of them I succeeded.


Please sign in to download the files of this thesis.