AWR1243BOOST: P Cascade Radar: object detected at wrong range and bad resolution

Hi, 

I am using the TI AWR1243P Cascade Radar evaluation board (TIDEP-01012) with the TI MMWCAS-DSP-EVM Evaluation module (TIDEP-01017).
Our results seem to indicate that our radar is not functioning as intended. The range values are wrong and we have no angle resolution. For example, when we place a small corner cube with metallic tape at 3 m, we record it at 13 m.  Also, the resolution of the sensor seems to be lower than we would expect.

Here are some of the results. We took data outside and placed a corner reflector covered with metallic tape at different distances in front of the radar . We use a MIMO script with a configuration that gives a maximum range of 15 m. The maximum range appears correct in the data but the range of the object is wrong. 

Here is the configuration we are using:
start_frequency = 77 GHz
slope = 79 MHz/us
idle_time = 5us
adc_start_time=6us
adc_samples=256
sample_freq=8000 ksps
ramp_end_time = 40us
rx_gain = 48dB

start_chirp_tx = 0 
end_chirp_tx = 11
nchrips_loops=64
nframes_master/slaves = 10 
Inter_Frame_Interval = 100

Setup image :


Result for the reflector at 1m. We see that it appears at about 5m. 

Result with the reflector at 2m. We see that it appears at about 8m. 

Results with reflector at 3m. We see that it appears at 13m . 



Result with the reflector at 4m. We see that it does not appear in the frame. 


In all cases, the angular resolution is very bad. 

We think we may not have done the calibration properly. For now, we’ve used the corner cube covered in metallic tape 5m in front of the sensor and captured the calibration files outside.  Could it be the case? 

Another idea was that maybe we haven't activated all the reflective antennas, that would explain the fact that we don't have any angular resolution. We are now using the lua scripts Cascade_Configuration_MIMO.lua and Cascade_Capture.lua within mmwavestudio to capture the data. Is there a way to disable/enable the Rx antennas to check?

We are happy to hear any ideas / suggestions of tests we could do to understand what is going on. 

Thank you very much, 

Alice