Channel Characterization for Intra-Wagon Communication at 60 and 300 GHz Bands

• Published in: IEEE transactions on vehicular technology Vol. 68; no. 6; pp. 5193 - 5207
• Main Authors: Guan, Ke, Peng, Bile, He, Danping, Eckhardt, Johannes M., Rey, Sebastian, Ai, Bo, Zhong, Zhangdui, Kurner, Thomas
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.06.2019; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Antenna measurements; Bandwidth; Channel characterization; Communications systems; Computer simulation; Delays; Elevation; Frequency measurement; High speed rail; intra-wagon; Line of sight; Loss measurement; Millimeter waves; mmWave; Rails; railway communications; ray-tracing; Reflection; smart rail mobility; Systems design; Three-dimensional displays; THz
• ISSN: 0018-9545; 1939-9359; 1939-9359
• DOI: 10.1109/TVT.2019.2907606

In this paper, the intra-wagon channels at 60 and 300 GHz bands are characterized through measurement-validated ray-tracing (RT) simulations. To begin with, an in-house-developed three-dimensional RT simulator is calibrated and validated by a series of millimeter-wave and Terahertz channel measurements inside a high-speed train wagon. Then, the validated RT simulator is used to conduct extensive simulations with different transmitter (Tx) and receiver deployments. At low frequencies, the channel is strongly influenced by the line of sight (LOS), and therefore, is usually classified into LOS and non-LOS (NLOS) regions. However, the simulation results at 60 and 300 GHz bands show that the first-order reflection also imposes a significant impact on the channel characteristics. This motivates us to further classify the NLOS region into light-NLOS (L-NLOS) and deep-NLOS (D-NLOS) according to the existence of the first-order reflection. Through analyzing the area ratios of LOS, L-NLOS, and D-NLOS regions, we evaluate the Tx deployment strategies and suggest the optimum one. Based on RT simulation results, totally 12 cases (three propagation regions with two Tx deployments at two frequencies) are characterized in terms of path loss, shadow fading, root-mean-square delay spread, Rician K-factor, azimuth/elevation angular spread of arrival/departure, cross-polarization ratio, and their cross correlations. All these parameters are fed into the 3GPP-like quasi-deterministic radio channel generator (QuaDRiGa). The good agreement between QuaDRiGa and RT proves that the 13 tables provided in this paper effectively parameterize the intra-wagon scenario for the standard channel model family. These results provide valuable insights into the system design and evaluation for intra-wagon communications.