You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I am referring to the following fragment of the Primer on Electromagnetics (and :
I have two problems with this description
Why is there no information about the phase shift resulting from $\boldsymbol{r}$? If $\boldsymbol{E_s}$ is an electromagnetic wave at the point $\boldsymbol{r}$ , then there should probably also be information about the phase shift $e^{2\pi jk|\boldsymbol{r}-\boldsymbol{q}|}$. Has this phase shift has been included in the end?
The diffusion model assumes that the phase shift resulting from reflection is sampled from the range
[0 $2\pi]$. Where does this assumption come from? Since we know that a rough surface consists of many rays reflected at different angles (and Path Solver finds them) knowing the angle between the incident and reflected wave and the material and its thickness, the phase shift can be estimated as $-\frac{h}{cos\phi_i}+\angle(r(\boldsymbol{\hat{k_i}}, \boldsymbol{\hat{k_s}}))< \chi < \frac{h}{cos\phi_i}+\angle(r(\boldsymbol{\hat{k_i}}, \boldsymbol{\hat{k_s}}))$. If we can sample from such a range, then for edge cases (wave incident perpendicularly, small material thickness) this should improve the ability to predict channel quality (CIR).
It would also be nice if the scattering characteristic could be complex. Even if this doesn't make much sense for diffuse reflection, when we're talking about discrete, small materials with known geometry, such characteristics can be determined (e.g., for a sphere).
The distance-based phase shifts are applied. They are simply missing in the equation. Will be fixed in the next release. Concerning the random phase shifts, this comes from Equation 9 in this paper. Not sure how 3) relates to diffuse reflections. I convert this to a discussion.
Activity
CSW-KR commentedon Sep 2, 2026
Hello sir could you already solve this problem?
jhoydis commentedon Sep 2, 2026
Hi @MieszkoP @CSW-KR,
The distance-based phase shifts are applied. They are simply missing in the equation. Will be fixed in the next release. Concerning the random phase shifts, this comes from Equation 9 in this paper. Not sure how 3) relates to diffuse reflections. I convert this to a discussion.