IWRL6432: Implementing AoA algorithm for separate vertical arrangement

Hi Wolfgang, 

I'm looking into your question. Please give me another day to get back to you. 

Thanks,

Josh

Sure Josh, No problem (y) 

Hi Wolfgang, 

I apologize for the delay here, I'm still looking to confirm exactly where changes would be required but I believe it will require more than just changing the antGeometryCfg commmand. 

regards,

Josh

Hi Josh, may I ask you if there is any update?

And how about this smaller arrangement of 3 vertical and 3 horizontal elements like bellow?

Thanks, 

Wolfgang

Wolfgang, 

Sorry about the delay. Here is an update on this. 

I'd first like to correct my earlier statement, for a custom antenna pattern such as the one you shared it is expected to work with the only expected modification being to the antGeometryCfg command.

I tried to confirm that this would be feasible by adding 'antGeometryCfg 2 0 1 0 1 1 1 2 0 2 0 3 2.5 2.5' to the configuration. 

Unfortunately this results in a error, so there is currently a bug with this functionality. For a bit more context, all the radar processing for IWRL6432 is done using the HWA which requires what we call param sets to configure how it operates. For this device there are a limited number of param sets that can be used and for this antenna pattern it requires more param sets to do the processing and it ends up being more than this DPU is allowed to use. This seems also to be the case for the other pattern you have shared as well. I've reported this to our software team so that they may be able to fix this in the future and to see if they have any quick workaround you can use. Will get back on this. Very sorry again about the delay. 

Best regards,

Josh

Thanks Josh for your (although la bit late but) insightful answer.   If I may, I would like to explore your answer about HWA param sets number and compare this antenna pattern to the wcsp antenna pattern as follow:

what I think happened in the either aoa2dprcoess or doaprcoces for azimuth/elevation angular process are as follow:
1. Azimuth fft processing for both rows.
2. Alignment of second row to the first row, due to both row differs in horizontal position by one wavelength.(This is my speculation because unlike step number 1 and 3, I can't see this process written any where in the code. Can you confirm or deny this step?  And can you point out where this is happened? Or it just simply done by vacated first two places of second row due to antenna mapping? )
2. Elevation fft from the result of above step.

My question are:
1.Confirmation of steps number two
2.It seems to us that above steps can be implemented in our antenna pattern without any additional param set. So why we need more param steps?
3. Is there any hard limitation of HWA params set number?

Thank you very much,

Sincerely,

Wolfgang

Hi Wolfgang, 

I apologize as I have not had a chance to review these latest questions. Please give me another day to respond here. Feel free to ping the thread if I have not replied in 24 hr.

Best regards,

Josh

Hi Josh,

May I poke you if there is any progress in this issue, or if you want to correcting my understanding?

Thank you very much,

Sincerely,

Wolfgang