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: A step or shift in received signal level from one burst to another

Part Number: MMWCAS-DSP-EVM
Other Parts Discussed in Thread: MMWCAS-RF-EVM

Hi,

I am using the MMWCAS-DSP-EVM + MMWCAS-RF-EVM kit.

I use the advance frame configuration to create the frames. The radar scans a sector from -60 to 60 degrees off-boresight in steps of 2 degrees, gives 61 directions or bursts.

I have used the same configuration for a while, and it has worked fine till now. Now I experience a shift or step in received energy between burst number (direction) 32 and 33 (See attached figure from Matlab). Frame based beam steering with 64 chirps in each burst. If I reduce the number of chirps to 32 in a burst it works fine, but it has worked with 64 formerly. What is the problem?

Here are the main parameters I use to configure the radar from a Matlab file. Tell me if you like the JSon files.

%% define what angles to steer in TX beamforming mode
anglesToSteer = -60:2:60; % angles to steer in TX beamforming mode, in unit of degrees. 0 is boresight, positive clockwise

%% chirp/profile parameters
start_Freq_GHz = 77; % starting frequency for chirp, make sure the entire BW is within 76~77 or 77~81
slope_MHzperus = 20; % MHz/us
idle_Time_us = 25; % us
tx_Start_Time_us = 0; % us
adc_Start_Time_us = 5; % us
ramp_End_Time_us = 35; % us
sampling_Rate_ksps = 10000; % ksps
samples_per_Chirp = 256; % Number of samples per chirp
rx_Gain_dB = 30; % dB
% Frame config
nchirp_loops = 64; % Number of chirps per frame
num_Frames = 100; % number of frames to collect data
framePeriod = 500; % ms, instead of defining duty cycle (ON duration)/(ON+OFF duration)

%advanced frame config
chirp_Frame_BF = 0; % 0 - frame based beam steering, 1 - chirp based beam steering, 2 - as 1, but a trick to handle long chirps (more than 255) in one direction
numSubFrames = 1; %Currently only 1 subframe is defined



  • Hi,

    It is difficult to root cause the problem based on your description.

    Do you think it is a sw or hw problem?

    Did you make any changes to the sw that could explain the change in behavior?

    Thank you

    Cesar

  • Hi Cesar,

    I understand it is difficult. I also find it quite frustrating. I can't see what I have changed that should cause this.

    I have done two things since it worked. I have tried to reduce times involved as much as possible to get minimum frame period betwwen two subsequent  angular sweeps, i.e. idle times and increased duty cycle. Typically when it fails you get symptoms like this one, but this configuration has worked before and why shouldn't it work again? The second thing is that I have downloaded mmWaveStudio 3.00.00.14, Matlab and so on, to another computer and uses this to configure and control the radar. It was when controlling the radar from this computer I experienced the change in behaviour. But I have returned to control the radar from my original computer and it still fails configuring the radar from this computer?

    Regards

    Aanund

  • Hi,

    I understand that this could be frustrating.

    If you modified the frame period betwwen two subsequent  angular sweeps, this could definitely result in a different behavior.

    You will need to continue to make experiments to return to a state where the use case is working.

    Based on your description it seems that the board is working fine.

    So the difference must be in the sw configuration

    thank you

    Cesar

  • Hi Cesar,

    I probably put it out unclear. A configuration that has worked before does not work now. I tried to explain what I have been doing with the radar from that time it worked until I realized it did not work for some configurations, that might have caused the problem. I don't change the frame period within one configuration. Different frame periods are uploaded as two different configurations.

    I will try to investigate it further.

    Regards

    Aanund

  • Thank you for the clarificatio

    Cesar

  • Hi,

    I believe what I experience may be caused by problems in the data recording? In chapter 11 and 12 in the mmWave Radar Interface Control Document quite a few timing and inter timing issues are described, especially if one uses the advanced frame configuration. It is not straight forward to realize what times should be added here and there... Are there similar limitations for the data recording, CSI Lane part? The odd thing is that the configuration has worked before.

    Regards

    Aanund

  • No, this should not be an issue with data capture. The sw should take care of it.

    thank you

    Cesar