IWR6843ISK: Range Azimuth Heat Map: How to parse TLV and compute data

Quentin BAJOT

Intellectual 480 points

Part Number: IWR6843ISK

Hello Support Team,

I'm using IWR6843ISK and Area Scanner project from SDK 3.1.1.2. And here is the cfg file I use :

I'd like to compute the same Azimuth-Range Heatmap as in OOB Demo.

I receive the TLV of the Azimuth-Range Heatmap but I have somme issues of understanding it and computing the data.

When I have a A-R HeatMap TLV (Type=4), the length of the TLV is equal to 4096. But using the formula of the document mmw Demo Data Structure v0.1 : Size= #RangeBins * #VirtualAntennas * 4 (bytes)

Because of my conf, I have 4 RX and 3 TX, so I would say 12 Virtual Antennas. But do I have to include the Elevation Antenna in the computation of the Virtual Antennas ?

And number of Range Bins is equal to the smallest power of 2 greater or equal to numADCSamples, so 512.

Therefore I should expect 512*4*12=24 576, so I think I've missed something.

Could you help me understand this difference and what are the next steps of processing those data? I'm currently working with the source code of the OOB Demo but I don't understand it very well.

Thanks a lot !

Quentin.

over 5 years ago

0 Former Member over 5 years ago

Hello Quentin,

Here is a code snippet to parse the heatmap

% range-azimuth static heat map, this is a 2D FFT array in range direction (x[numRangeBins][numVirtualAntAzim]), at doppler index 0, sized to azimuthStaticHeatMapSize elements of type cplx16ImRe_t
function [Q, Qcorr, q] = getAzimuthStaticHeatMap(payload, numTxAzimAnt, numRxAnt, numRangeBins, numAngleBins, nearEndCorr)
    [r c] = size(payload);
    if(r<c)
        payload = payload';
    end
    
    numVirtualAntenna = numTxAzimAnt * numRxAnt;
    q = payload;
    q = q(1:2:end)+q(2:2:end)*256;
    q(q>32767) = q(q>32767) - 65536;
    q = 1j*q(1:2:end)+q(2:2:end); %Imaginary first, real second
    q = reshape(q, numVirtualAntenna, numRangeBins);
    

    Q = fft(q, numAngleBins);

    %Near field corrected
    q1=q;
    q2=q;
    q1(5:8,:) = 0;
    q2(1:4,:) = 0;
    Q1 = fft(q1, numAngleBins);
    Q2 = fft(q2, numAngleBins);
    Qcorr = Q1 + Q2.* nearEndCorr;

return

Best,

Amanda

0 Quentin BAJOT over 5 years ago in reply to Former Member

Intellectual 480 points

Hello Amanda,

Thanks a lot for this code snippet !

I test it and I come back to give you feedback,

Have a great day,

Quentin.

0 Quentin BAJOT over 5 years ago in reply to Quentin BAJOT

Intellectual 480 points

Hello again !

I managed to add the code snippet in my visualizer. I used numRangeBins=256, numAngleBins=64 but I don't know why those parameters are working in theory.

After computing magnitude of the Q matrix, I get the following HeatMap:

Am I write saying we can see here what each RX antena "receive" ?

From those data, could I create a HeatMap like in OOB Demo ? (in polar coordinates, or not, but global)

Thanks for your help !

Quentin.

0 Former Member over 5 years ago in reply to Quentin BAJOT

Hello Quentin,

% plot single frame of azimuth static heatmap
if(~isempty(frame(frameIdx).azHeatmap))
    figure('Name','Azimuth Static Heatmap')
    Q = frame(frameIdx).azHeatmap.Q;
    NUM_ANGLE_BINS = size(Q,1);
    theta = asind([-NUM_ANGLE_BINS/2+1 : NUM_ANGLE_BINS/2-1]'*(2/NUM_ANGLE_BINS));
    range = rangeBin2Meters;
    posX = range' * sind(theta');
    posY = range' * cosd(theta');
    xlin=linspace(-floor(range(end)),ceil(range(end)),200);
    ylin=linspace(0,ceil(range(end)),200);
    [X,Y]=meshgrid(xlin,ylin);

    QQ = fftshift(abs(Q),1);
    QQ=QQ.';
    QQ=QQ(:,2:end);
    Z=griddata(posX,posY,QQ,X,Y,'linear');
    hAzimuthHeatmapPlot = surf(xlin,ylin,Z);
    shading INTERP
    view([0,90])
    axis('equal')
    axis([-ceil(range(end)) ceil(range(end)) 0 ceil(range(end))])
    xlabel('Range X-Axis [m]');                  
    ylabel('Range Y-Axis [m]');
    titleStr = sprintf('Frame %d, Azimuth Static Heatmap', frame(frameIdx).header.frameNumber);
    title(titleStr)
end

Here's a snippet to plot a similar heatmap as in OOB.

Amanda

0 Quentin BAJOT over 5 years ago in reply to Former Member

Intellectual 480 points

Thanks a lot Amanda,

It works perfectly !

Have a nice day,

Quentin.

Sensors

Sensors forum

IWR6843ISK: Range Azimuth Heat Map: How to parse TLV and compute data