3-D Geometry-Based Statistical Modeling of Cross-Polarization Discrimination in Wireless Communication Channels

Abstract

A new three-dimensional (3-D) geometry-based reference model for cross-polarization discrimination (XPD) in narrow-band fixed-to-fixed (F-to-F) or fixed-to-mobile (F-to-M) wireless channels is proposed. The proposed model is based on the superposition of polarization components on conservation-ofpolarization planes (CoP-planes). This novel approach is used to derive the XPD at each distance without the aid of measurement data or its approximation. To the best of our knowledge this is the first purely theoretical geometry-based XPD reference model. Using our reference model, we obtain a relationship between XPD and channel modeling factors such as the distance between the transmitter and the receiver, the azimuth/elevation angles of arrival and departure (A/E-AoA and A/E-AoD), multipath delays, and the distribution of scatterers. Our analytical XPD model is in close agreement with previously reported results from other authors that are based on empirical measurement data, and it also overcomes the shortcoming of overapproximation/oversimplification of the measurement data.

Cross-Polarization Discrimination in Vehicle-to-Vehicle Channels: Geometry-Based Statistical Modeling

“…However, the XP D V V/HV of the SBR-only propagation with appropriate model parameters is in substantially close agreement with the empirical data of the F2F channels in [6] or the F2F/F2M reference model in [10], especially when D ranges from 2 km to 10 km. The reason is that [6] and [10] take into account the highelevation Tx antenna and low-elevation Rx antenna so that only the SBR propagation happens. The SBT propagation can be considered as a symmetry for the SBR as also indicated in [12], and thus applying our principal XPD modeling method, the superposition of the CoP-planes' polarization components, to the SBT propagation is reasonable.…”

“…3, the modeling parameters have been achieved such that the total XPD value considering both the SBT and SBR approaches 0 dB. However, the XP D V V/HV of the SBR-only propagation with appropriate model parameters is in substantially close agreement with the empirical data of the F2F channels in [6] or the F2F/F2M reference model in [10], especially when D ranges from 2 km to 10 km. The reason is that [6] and [10] take into account the highelevation Tx antenna and low-elevation Rx antenna so that only the SBR propagation happens.…”

“…Then the angle between V' SBT and its projection line on the same plane is also φ V [13]. For this reason, we call this Tx-scattererRx plane a conservation-of-polarization plane (CoP-plane) [10]. The polarization vector, V' SBT on its CoP-plane can be decomposed into the vertical and horizontal polarization components at the Rx.…”

“…Here, f c and f D are, respectively, carrier and Doppler frequency, and case ∈ {(SBT, 1), (SBT, 2), (SBR)} respectively. For A HV , A HH , and (26) is replaced with f HV , f HH , and f V H , respectively, and A H is substituted for A V in the case of [10], but they are not re-written in this paper.…”

“…Correspondingly, radio propagation is assumed to be conducted in an out-door suburban or metropolitan environment that is characterized by 3-D non-isotropic scattering with non-line-of-sight (NLoS) conditions between the Tx and Rx. The fundamental principle for statistical modeling of the XPD in V2V channels in this paper is the superposition of the polarization components on CoP-planes, an approach that we have proposed previously for F2F/F2M channels with high-elevation Tx antennas [10].…”