What is the axis for V-polarization and what is the axis for H-polarization ?

[golnazsalehi]

Based on the docs, a planar array is defined in the Y-Z plane and V-polarization means along the Z-axis and H-polarization means along the Y-axis. So a H-polarization dipole antenna should have a null on the Y-axis.

So I defined two 1x1 planar arrays (one Tx, one Rx), both H-pol with dipole pattern, and I put them on the Z-axis (meaning that the location of one of them is [0,0,1] and the other on is [0,0,0]). The calculated channel for all of the subcarriers is zero!

I put them on the Y-axis and repeated the simulation, I got zero for all channels!! Why is that?

When I change the pattern to 'iso' it makes sense meaning that I get zero channels when I put them along Y-axis and non-zero channels along any other axis.

[jhoydis]

Hi @golnazsalehi,

I think that there is a misunderstanding on your side with respect to the meaning of V-/H-polarization.

We typically deal with spherical waves and the electromagnetic field components must be orthogonal to the propagation direction, which is described by the radial vector $\hat{\mathbf{r}}$. V-polarization means that the field is aligned to the elevation direction $\hat{\boldsymbol{\theta}}$, and H-polarization means that it is aligned to the azimuth direction $\hat{\boldsymbol{\varphi}}$ (as defined in (1)). These directions are dependent on the elevation and azimuth angles of the radial vector.

Concerning your first experiment, please note that the dipole antenna pattern is zero for $\theta=[0,\pi]$. It is therefore normal that you do not receive any field.

Concerning the second experiment where the locations are [0,0,0] and [0,1,0], you should get a non-zero line-of-sight channel if both antennas have the same polarization and are not rotated in any way.

[golnazsalehi]

Hi,
Would you please answer the following question:

If I have a V-pol/H-pol dipole antenna, the field is aligned with which axis? Based on the attached files the answer is Z-axis/Y-axis.
Am I wrong?

[jhoydis]

Hi @golnazsalehi,

The visualization does not correspond to the orientation of the electric field the antennas radiate.

As I have tried to explain above, a horizontally polarized antenna radiates an electric field whose field vector is aligned with the spherical azimuth vector. This vector coincides with the +Y-axis for $\theta=\pi/2$ and $\varphi=[0]$.

[golnazsalehi]

Hi
does Sionna consider depolarization as well?
I mean if the angle of incidence is non zero (zero is normal incidence) on a surface, would the reflected ray has a different polarization as the incident ray?

[jhoydis]

Yes, it does. Please refer to the section on Reflection and Refraction as well as the section on Scattering. For diffuse reflections, you need to assign a cross-polarization discrimination coefficient to a radio material that determines how much energy will be transferred from one polarization direction into the other.

[golnazsalehi]

Thank you for your respond.
Is there a doc of how to assign the xpn-coef for different material for different frequencies?

[golnazsalehi]

Hi again
I change the xpd_coefficient manually and it would make no difference in the channels between the Tx and Rx.
Is this coefficient trainable?

[jhoydis]

Yes, it is trainable.
Please have a look at the ray tracing tutorials. and the API Documentation. Everything is explained.