[Solved] O1 scenario, channel frequency response
I'm working on a SISO OFDM configuration in the O1 scenario at 60GHz, and I am having trouble with some results.
- I don't understand how the channel model (Equation 1 in the DeepMIMO paper on Arxiv) represents the channel at a given subcarrier frequency. In this Equation, if we consider that the index k represents the indices of the subcarriers (array starting at 0 and finishing at the wanted number of subcarriers), we obtain an array starting at 0 and finishing at the bandwidth frequency B (with the multiplication of k/K with B). But this is not at the central frequency (here 60 GHz). Is the information about the RF central frequency contained in the path phase ?
- If we consider only the BS 16, there are some UE positions that the dataset consider as being LoS that are not if we look at the O1 map (for example for the UE row 1, cf. attached file).
Thank you in advance for your help !
-The scenarios are generated with the ray-tracing at the RF central frequency. Therefore, the path parameters (including the phase) reflect that.
-Thanks for the question. There was a small bug in the LoS status labels causing this mismatch for some users. We have fixed it and updated the scenario files. You can download it from the website again. It will now be correct.
Thanks for your response.
- Can you be a little bit more specific about your answer ? In Equation (1) on the Arxiv paper, we could think that the path phase corresponds to a frequency transposition term (2 pi f_RF tau), where tau is the path delay. However if we access the path phase value in debug mode, we don't obtain this value (2 pi f_RF tau). I don't really understand how the RF transposition is done. I hope you can enlighten me.
- Regarding the LoS/NLoS bug, I just tried with the new scenario files and it is now correct, thank you very much.
Hi Baptiste (@bchateli),
The phase values of the different paths are calculated by the Wireless Insite ray-tracing software (accounting for the operating frequency, scatterers materials, etc.). The DeepMIMO generated channels account for that phase using the attached equation.