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.

IWR6843: Rx Phase information from RAW data

Part Number: IWR6843

Hi


I am trying to get some information for the phase of each RX antenna.


The setup I am using is as follows:

I have one IWR6843 (call it transmitter) that transmits a 61.25GHz CW wave.

I have one IWR6843 (call it receiver that transmits a 61.255625GHz CW wave. (5.625MHz offset from the transmitter)

In the transmitter I collect the data using a DCA1000.

The transmitter and receiver is put in front of each other at a distance of about 1.5m

We move them relative to each other between measurements so that the angles between the receivers Rx's and the transmitters Tx is somewhere from 0 to +/- 180 degrees.

We have implemented the software needed to get this to work.


The data we get regarding the phase of the different RX antennas in the receiver looks odd tough.

We extract the phase information as follows:


Run a fft over the raw data cube after applying a window function, (Python)

       r_win = np.blackman( 128 )

       r_cube = np.fft.fft(cube * win.reshape(-1, 1, 1), axis=0, 128 )

Then we loop over the r_cube and finds the FFT-bin for the peak at 5.625MHz.

That bin is then used to extract the phase and amplitude for each Rx by calling:

cmath.polar(x) where x is the value from that FFT-bin for each Rx.


Since the phase at a given moment in time is random, but the relative phase between the Rx:s  is not, we subtract the phase from Rx0 from all the other Rx1-3.

We convert the phase to degrees and get the results.


Things that are odd:

- The phase shifts from positive to negative from time to time during a measurement.

- All Rx have similar relative phases but they all vary a lot even when the transmitter and receiver is not moved and are standing facing each other at a distance of about 1.5m

My questions: Can you tell me how to extract the relative phase information between Rx:s?

Why dies the value of the relative phases differ from -180 to +180, they seem to follow some type of pattern, where they move from smaller to bigger and back again, ignoring the sign.

Result from measurement:

Relative phase: <Rx0_1> <Rx1_2> <Rx2_3>

Relative phase: -82.48 -97.08 -88.52
Relative phase: -104.38 -106.83 -94.43
Relative phase: -109.12 -97.28 -102.84
Relative phase: -123.38 -118.87 -125.82
Relative phase: -142.24 -120.65 -143.29
Relative phase: -137.18 -141.33 -140.56
Relative phase: -147.91 -145.63 -158.13
Relative phase: -149.22 -157.25 -165.51
Relative phase: -153.67 -172.04 -152.4
Relative phase: -173.32 -164.1 -150.7
Relative phase: -165.95 -176.28 -155.3
Relative phase: 168.33 180.63 126.34
Relative phase: -164.25 -190.43 -175.38
Relative phase: 161.63 175.43 143.29
Relative phase: 141.38 143.52 146.97
Relative phase: 138.13 131.39 144.17
Relative phase: 122.04 132.76 140.51
Relative phase: 122.1 146.44 129.2
Relative phase: 133.93 120.35 127.35
Relative phase: 109.72 116.6 106.97
Relative phase: 108.78 95.23 110.02
Relative phase: 94.55 98.97 91.69
Relative phase: 90.35 107.02 93.46
Relative phase: 95.57 99.24 98.56
Relative phase: 83.46 95.08 85.49
Relative phase: 99.52 83.57 86.89
Relative phase: 91.37 74.35 96.9
Relative phase: 81.03 91.96 65.8
Relative phase: 67.01 82.98 73.79
Relative phase: 70.31 70.43 68.98
Relative phase: 68.99 64.28 61.27
Relative phase: 82.18 66.43 55.6
Relative phase: 71.78 54.13 67.74
Relative phase: 31.98 84.25 37.26
Relative phase: 51.58 50.67 35.95
Relative phase: 59.74 37.28 39.85
Relative phase: 44.34 40.47 38.15
Relative phase: 22.04 29.89 26.97
Relative phase: 43.08 20.84 15.68
Relative phase: 46.53 5.26 33.56
Relative phase: 20.43 24.37 15.63
Relative phase: -0.73 -331.66 -342.46
Relative phase: 10.59 4.44 3.23
Relative phase: -9.07 -0.02 -351.1
Relative phase: 7.66 350.3 358.77
Relative phase: -6.6 -13.33 -0.86
Relative phase: -13.9 -26.57 -22.17
Relative phase: -12.69 -31.0 -42.23
Relative phase: -42.68 -27.88 -19.08
Relative phase: -46.22 -26.84 -39.94
Relative phase: -31.57 -35.96 -29.43
Relative phase: -39.88 -31.49 -46.08
Relative phase: -57.35 -26.31 -41.87
Relative phase: -65.84 -38.94 -44.19
Relative phase: -49.77 -59.68 -46.81
Relative phase: -58.64 -46.9 -57.23
Relative phase: -74.45 -54.7 -50.71
Relative phase: -54.69 -65.1 -60.28
Relative phase: -66.6 -69.27 -80.27
Relative phase: -78.27 -64.23 -79.91
Relative phase: -72.38 -76.16 -75.99
Relative phase: -89.52 -76.12 -82.05
Relative phase: -78.36 -89.6 -94.44
Relative phase: -72.89 -98.9 -84.16
Relative phase: -99.17 -85.46 -93.97
Relative phase: -107.11 -100.7 -88.7
Relative phase: -108.71 -95.32 -90.22
Relative phase: -86.69 -120.43 -90.06
Relative phase: -105.93 -114.51 -103.85
Relative phase: -108.81 -123.18 -105.21
Relative phase: -106.45 -121.01 -115.4
Relative phase: -126.95 -122.87 -103.57
Relative phase: -129.18 -120.38 -119.71
Relative phase: -140.91 -119.35 -131.96
Relative phase: -144.47 -117.06 -144.55
Relative phase: -139.4 -123.07 -156.89
Relative phase: -137.33 -160.7 -150.8
Relative phase: -144.55 -152.55 -146.41
Relative phase: -146.94 -150.91 -154.55
Relative phase: -150.6 -150.5 -161.76
Relative phase: -165.16 -158.82 -151.51
Relative phase: -156.64 -163.8 -152.62
Relative phase: -164.16 -143.51 -179.06
Relative phase: -174.5 -164.03 -167.64
Relative phase: -153.98 -176.74 -175.5
Relative phase: 158.26 178.74 127.48
Relative phase: 171.72 162.92 181.14
Relative phase: 152.94 159.95 155.79

  • First, we never tried your setup.  

    You mentioned that at the receive side, you finds the FFT-bin for the peak at 5.625MHz.    Do you always see the peak at 5.625MHz?  Since the transmit and receiver has independent LO, the frequency can be off quite a lot between the the transmit and receiver. 

    I would need to check whether the 4 RX channel are sampled at exactly same moment. 

    Best,

    Zigang

  • It would be interesting to know if the Rx channels are sampled at the same time, please do check if possible, if that is the case.

  • This seems to state that all the Rx channels are samples simultaneously:

    There is also the strange behavior that some of the Rx channels end up with no peak in the FFT sometimes. Sometimes I can see all Rx channels not detecting any peak after the FFT.