This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

MMWCAS-DSP-EVM: Range / Doppler Issue with provided Matlab Scripts


I am using the MMWCAS-DSP-EVM, capturing data and post-processing it with the provided Matlab scripts. The board was calibrated using a corner reflector at 2m with the included cascade MIMO configuration (Lua script) and angular resolution and range resolution were verified. The image below was collected using the following parameters:

start_freq = 77 -- GHz
slope = 5 -- MHz/us
idle_time = 40 -- us
adc_start_time = 15 -- us
adc_samples = 256 -- Number of samples per chirp
sample_freq = 5000 -- ksps
ramp_end_time = 80 -- us
rx_gain = 48 -- dB

I have three questions from the data:

1) What could be causing the periodic doppler? This is most notable in longer range waveforms (>10m maximum range).

2) There is constructive and destructive interference in the range FFT, most observable in the range-azimuth plot as alternating concentric rings and in the range profile plot below. This phenomenon is in every data collection (see image below collected using included MIMO configuration) I have worked back to the adcData matrix in the Matlab processing code and still see the interference pattern.

One more question, I have had to scale the range values by 4.41. When calibrating I couldn't see targets at the recommended 5m range using a configuration that, according to the Matlab code, should have 13 meter range. Placing the target at 2m produced a return visible at 8.82 meters. The data collection was performed using the included MIMO Lua script. Do you have an explanation for that error?

My suspicion for the doppler issue is that when the adcData matrixes are being grabbed from the raw data files there is data that is being skipped, causing an offset of detected peaks. That explanation could explain the interference pattern in range, too, but I haven't found any bugs as I've been digging through the post processing code.