Search

Analysis and modeling of wireless communication systems are dependent on the validity of the wide-sense stationarity uncorrelated scattering (WSSUS) assumption. However, in high-mobility scenarios, the WSSUS assumption is approximately fulfilled just over a short time period. This paper focuses on the stationarity evaluation of high-mobility multi-band channels. We evaluate the stationarity time, the time over which WSSUS is fulfilled approximately. The investigation is performed over real, measured high-mobility channels for two frequency bands, 2.55 and 25.5 GHz. Furthermore, we demonstrate the influence of the user velocity on the stationarity time. We show that the stationarity time decreases with increased relative velocity between the transmitter and the receiver. Furthermore, we show the similarity of the stationarity regions between sub-6 GHz and mmWave channels. Finally, we demonstrate that the sub-6 GHz channels are characterized by longer stationarity time., Comment: Accepted at the XXXVth URSI General Assembly and Scientific Symposium (URSI GASS 2023)

In vehicular environments, the scattering environment can change very rapidly, owing to the mobility of the Tx and Rx vehicles, the relatively low height of the antennas and the potentially large number of scatterers in the surrounding environment. Vehicular radio propagation channels thus differ significantly from those of other wireless cellular networks, so that dedicated measurement campaigns and realistic radio propagation channel models are required to fully exploit their potentialities. The main contributions of this thesis are summarized as follows: - Two measurement campaigns of MIMO vehicular radio propagation channels were conducted at 5.3 GHz, - The validity of the WSS and US assumptions is assessed using the CMD, which characterizes the amount of change in the spatial structure of MIMO radio propagation channels. Stationarity distances and frequency bandwidth are 10 to 40 m long and 40 to 50 MHz wide, respectively, depending on the variability and richness of the scattering environment, - The non-stationary fading statistics are characterized stochastically, in terms of Ricean K-factor and RMS delay spread. Their time- and space-variations are strongly related to changes in the richness of the scattering environment and on the presence (or not) of a strong LoS component. They are modeled by multi-modal Gaussian distributions, - A vehicular GSCM is proposed, in which each scatterer is defined and characterized separately: the radio wave propagation with simplified ray-tracing tools and the scattering aspects with a classical fading model. The diffuse scattering is modeled in a purely stochastic manner. The GSCM is then completely parameterized and validated for SISO systems. (FSA - Sciences de l'ingénieur) -- UCL, 2013