Other Parts Discussed in Thread: DCA1000EVM,
Hi,
I have been trying to configure my AWR1642boost+DCA1000evm setup to BPM MIMO mode to obtain 2TX & 4RX (a virtual array of 8 receivers) and therefore a better angular resolution. So far, I was able to use my setup with 1TX with no problem but having some issues to understand the 2TX (BPM) mode configuration. I have read TI's MIMO Radar document , SDK document and bunch of threads. Eventually, these are the steps of how I configure mmWave Studio parameters for raw ADC data acquisition in BPM mode:
1. Connection setup.
2. Under SensorConfig Tab, in Profile section choose a profile ID (I believe single profile ID is enough for BPM mode and it can be used by different chirps) and set parameters.
3. In Chirp section, create 2 chirps. First one is: Profile ID=0, Start Chirp=0, End Chirp=0, both TX0 and TX1 are checked (for simultaneous chirp sending) and Set; Second chirp is: Profile ID=0, Start Chirp=1, End Chirp=1, both TX0 and TX1 are checked again and Set.
4. Go into BPMConfig tab. Go into BPM Chirp Config section. For the first chirp (in the MIMO Radar document (Section 4.2) it is referred as S_a=S1+S2): Start index=0, End index=0, TX0 BPM Val (TX On)=0, TX1 BPM Val (TX On)=0 and Set. For the second chirp (in the MIMO Radar document it is referred as S_b=S1-S2): Start index=1, End index=1, TX0 BPM Val (TX On)=0, TX1 BPM Val (TX On)=1 and Set.
5. Under Sensor Config tab go into Frame section. Set Start Chirp TX=0, End Chirp TX=1 and other parameters.
6. DCA arm +Trigger frame to collect data.
So, my questions are as follows:
I) Is there anything wrong with this configuration or is it the right way of configuring BPM MIMO parameters?
II) In raw data processing, according to the mimo radar document, we obtain s1 by averaging consecutive chirps ((S_a+S_b)/2) and s2 by ((S_a - S_b)/2). In this case, from 2 consecutive chirps (S_a and S_b) we get one s1 and one s2 but we should obtain two chirps from each of them according to the figure I believe, because simultaneous chirp transmission is the main advantage of BPM. So how do we obtain a total of 4 chirps out of S_a and S_b? Should we repeat s1 and s2 after obtaining them? This is what I do right now, I am replicating the chirps (s1 and s2) after obtaining them.
III) Do we need to do any further phase correction on chirps since we have sent S_b with 180 degree phase difference? If so, in which step? I guess it is recommended to do it after 2D-FFT (range-velocity/doppler map) but I am not really sure if it is a must because I can obtain micro-Doppler spectrograms without any phase correction. If it is a must for angle estimation, then it would be great to have some documents or guide to see how it is done.
IV) How should i reshape my data after obtaining my radar data cubes for 2 TXs to get 8 channel data cube (ADC samples, # of chirps, # of RXs)? Can I just concatenate data cubes two TXs? If so, in which order like concat(TX0,TX1) or concat(TX1,TX0)?
Thanks in advance for your help!
Best,
Emre
