Evaluation of intelligent surfaces in railroad communication

Abstract

Public transport is gaining more and more popularity throughout the world. As a consequence, the amount of users connected to the mobile network while traveling in trains keeps rising. Each user has different needs considering data rates, latency and reliability while traveling eco-friendly for several thousand kilometers per year in many places across the world. Providing the best possible quality of service to the aforementioned properties is the future of wireless networks beyond 5th generation (5G). Enabling all of these features is a challenging task, especially for railroad communications. This field of wireless communications has to deal with large distances, high velocity and many users, simultaneously. A potential candidate for the upcoming 6th generation (6G) network devices are Reconfigurable Intelligent Surfaces (RISs). A RIS consists of an array with electromagnetically reconfigurable elements in order to modify the response to incident signals. In this thesis, several use cases for intelligent surfaces in a railroad environment are discussed. A very promising case of application is to use a RIS to overcome Non-Line-Of-Sight (NLOS) situations. At first, this scenario is evaluated in terms of link level simulations. A power scaling law, impact of the K-factor of the employed Rician channel model and different imperfections are taken into consideration. After studying this topic, system level simulations are used to evaluate the impact of a RIS, if it has to cover an entire area. The scenario is rebuilt with realistic geometrical information from a railroad scenario in Austria. A major contribution is the analysis of different placement strategies for a RIS as well as the impact of two beamforming schemes. At the end of the thesis, a concluding example with a noticeable performance gain through a RIS is presented.