Other Parts Discussed in Thread: IWR1443, IWR1642
Hello,
currently I am working on building a MatLab raw data post processing chain, which can then be ported to the IWR in equivalent C code.
In the first catch I will build it like the actual processing / data path of the mmWave demo.
I have some trouble understanding the actual CFAR algorithm implemented on the HWA.
My understanding (based on for example the "MIMO Radar" paper ) is, that the CFAR should identify peaks on the noncohrenntly summed 2D FFT (of all virtual antennas).
In the mmWave doxygen documentation it seems like the CFAR gets the 2D FFT - so far so good.
But in the SDK user guide is noted, that for the IWR1443 "only Range direction is supported".
Does this mean that the CFAR is based on the 1D FFT of one chirp?
Or does it mean, that the CFAR checks each doppler bin in range direction?
In the doxygen documentation for IWR1642 it is a little bit more detailed. It says, that CFAR does a pre-detection in doppler direction and then final CFAR in range direction.
Could you please give some more detailed information about the CFAR implementation on HWA?
Is there a detailed documentation for HWA?
Thank you very much,
Enric
