3D Urban Channel Characterization for V2X Communications: Measurement-Driven Modeling of V2V and U2V Propagations

As the demands for high reliability, low latency, and high capacity in 5G/6G networks continue to grow, precise characterization of urban wireless channels has become essential. This study employs vector signal transceivers and Universal Software Radio Peripheral (USRP) devices for vehicle-to-vehicle (V2V) single-input single-output (SISO) and unmanned aerial vehicle-to-vehicle (U2V) multiple-input multiple-output (MIMO) scenarios. It establishes corresponding channel models and analyzes key channel parameters, including power delay profile, delay spread, angle of arrival, and Doppler shift. For the V2V SISO case, both a Gaussian model based on the Lp-norm and a non-Gaussian model based on the Spherically Invariant Random Process (SIRP) are developed. For the U2V MIMO case, a measurement-driven stochastic model with irregular geometry is proposed, which combines dynamic cluster birth and death behavior with multipath parameter randomization. The experimental results demonstrate that the Doppler power spectral density (DPSD), autocorrelation function (ACF), spatial cross-correlation function (SCF) and cumulative distribution function (CDF) of the root mean square delay spread (RMS-DS) exhibit excellent agreement with measured data, thereby validating the accuracy of the proposed channel model.