• State Resolved
  • Locked Locked
  • Replies 9 replies
  • Answers 1 answer
  • Subscribers 54 subscribers
  • Views 911 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

Original question:

CC2640R2F: IQ data from BOOSTXL-AoA is mixed up occasionally

CC2640R2F: Why is n't the phase different fixed between first sample of two antennas?

Andy Yang78
Intellectual 380 points
Part Number: CC2640R2F

Hello

We are using CC2640R2F to collect the I/Q data.

the following table shows the phase difference between 1st antenna 1st sample and 2nd antenna 1st sample.

Since the location of both TI transmitter and receiver are fixed and the AoA should be same.

Why the value of  phase different (Pab-Paa/3) isn't a fixed value?

http://dev.ti.com/tirex/content/simplelink_cc2640r2_sdk_2_30_00_28/docs/blestack/ble_user_guide/html/localization/index.html#angle-of-arrival

versus_avg[a][b] += Pab_rel - ((Paa_rel * abs(a-b)) /numAnt);

1 Phase A1.1 S0 Phase A1.2 S0 Pab  Paa Pab-Paa/3
2 -12.28666184 -4.136830871 8.149830967 　 　
3 -32.98852161 -28.44292862 4.545592989 -20.70185978 11.44621291
4 -59.70685047 -58.36933859 1.337511879 -26.71832885 10.2436215
5 -83.02238151 -88.77367214 -5.751290631 -23.31553104 2.02055305
6 -108.1189827 -113.0824663 -4.963483618 -25.09660114 3.402050097
7 -136.7936449 -136.9834767 -0.189831759 -28.67466228 9.368389001
8 -176.6236317 -164.6693928 11.95423888 -39.82998678 25.23090114
9 175.0816193 -178.7623208 -353.8439401 351.705251 -471.0790237
10 133.0666993 149.4817693 16.41507002 -42.01491998 30.42004334
11 120.364271 120.1317186 -0.232552359 -12.7024283 4.00159041
  • 0
    TI__Guru 57865 points
    Hi Andy,
    Shooting wildly here but isn't the signal traveling difference distances to the two antennas? (Ref. Trigonometry). There would also be some multi-path impacts present.

    Either way, assigning an expert on the topic.
  • 0 in reply to J Lindh
    Intellectual 380 points
    Thank you for the reply
    We collect the I/Q data in the chamber, should minimum the impact from multi-path.
    Is there any known cause for this phenomenon? such as the deviation from sampling, crystal or frequency drift.
  • 0 in reply to Andy Yang78
    Intellectual 380 points
    Dear J Lindh
    Is any expert able to support us on this topic? Thanks
  • 0 in reply to Andy Yang78
    TI__Genius 9215 points
    Hi Andy,

    What SDK are you using? I'm asking because first valid sample have changed between the different version.
    Can you provide the complete IQ raw data, not just 10 samples?

    Btw, the calculation in row 9 is wrong. The phase difference should be 6.16 degrees (not -353).


    Best Regards,
    R.M
  • 0 in reply to R.M
    Intellectual 380 points

    Hello R.M

    We are using SDK version 2.30.00.28.

    Btw, the calculation in row 9 is wrong. The phase difference should be 6.16 degrees (not -353). 

    ==>Yes, we notices row 9 data A1.1 and A1.2 cross two quadrants, the calculation should be modified.

    Except row 9,  other Pab/Paa isn't a near fixed value.

    Please refer to the following attachment for the IQ raw data, including ordered and distorted samples.

    IQ_data.zip

    Moreover, we also observed the distorted sample as following IQ diagram. 

    Therefore, we have two issues here.

    1. For ordered samples, the Pab isn't a near fixed value (measured in chamber).

    2. For distorted samples, the iQ data blow up, around 5 to 10% chance. Is there any known cause for this phenomenon．

    It will be nice if you can also check the 2nd issue for us.

    Regards,

    Andy

  • 0 in reply to Andy Yang78
    TI__Guru 56215 points

    Hi,

    We already have another expert helping you on issue 2 in this post

    so I will leave it to my colleagues.

    I did a quick calculation from first 20 samples R1 in your excel sheet TI_shielding_chamber.

    I am not able to get the values you show here, I am getting about 20 degrees of Pab for the first 9 samples from your excel.

    Can you show us how you convert the I/Q data to angle and then how you calculate the Pab?

    BTW, next time if you want us to take a look at the data, please save it either 1 column I, 1 column Q or 1 row I, 1 row Q.

  • 0 in reply to Christin Lee
    Intellectual 380 points

    Dear Christin

    I use same attached IQ data to calculate following Pab.

    I am using excel ATAN2(I, Q)*180/PI() to get the phase for each sample.

    Is the Pab output data reasonable?

    A1.1.0 A1.1.1 A1.1.2 A1.1.3 A1.1.4 A1.1.5 A1.1.6 A1.1.7 A1.2.0 A1.2.1 A1.2.3 A1.2.4 A1.2.5 A1.2.6 A1.2.7
    -125.747 -105.136 -83.5584 -58.7995 -37.3733 -16.2602 8.587936 31.20051 -107.368 -86.7928 -62.6501 -40.7374 -18.5138 5.957131 27.56236 49.42583
    -137.085 -116.058 -94.3635 -69.58 -47.6476 -26.1506 -2.74809 18.21982 -116.455 -95.1944 -71.1753 -50.3828 -29.6237 -8.61565 15.47133 37.4631
    -147.529 -123.818 -102.629 -80.8078 -56.3751 -34.5326 -12.9774 10.91113 -123.357 -101.549 -79.6513 -55.4915 -34.1013 -12.4751 11.55458 33.20275
    -148.613 -126.461 -106.133 -85.6882 -63.1329 -43.4 -22.7782 0.916654 -134.318 -113.037 -90.7538 -64.9067 -42.7094 -21.666 -0.95884 22.38014
    -148.715 -127.427 -103.867 -83.2221 -62.7071 -39.2404 -17.7004 4.610649 -132.274 -108.591 -87.2619 -66.0877 -42.8919 -22.4477 -1.7357 22.12381
    -159.661 -137.588 -112.778 -90.4584 -68.0704 -43.2152 -21.3335 0.684778 -139.918 -116.565 -95.6136 -74.7235 -51.2258 -30.922 -10.2614 10.58957
    -165.34 -144.142 -123.69 -100.731 -80.3879 -59.7436 -35.8064 -13.4778 -157.842 -137.579 -113.813 -92.1932 -70.6533 -46.3479 -24.7416 -2.91084

    Pab 0 Pab 1 Pab 2 Pab 3 Pab 4 Pab 5 Pab 6 Pab 7
    -18.3782 -18.343 -20.9083 -18.0621 -18.8595 -22.2173 -18.9744 -18.2253
    -20.63019 -20.8636 -23.1882 -19.1972 -18.0239 -17.535 -18.2194 -19.2433
    -24.17224 -22.269 -22.9781 -25.3163 -22.2738 -22.0575 -24.532 -22.2916
    -14.29504 -13.4245 -15.379 -20.7816 -20.4235 -21.734 -21.8194 -21.4635
    -16.44112 -18.836 -16.605 -17.1344 -19.8152 -16.7926 -15.9647 -17.5132
    -19.7424 -21.0228 -17.1646 -15.7348 -16.8446 -12.2931 -11.0722 -9.90479
    -7.497878 -6.56249 -9.8775 -8.53797 -9.73463 -13.3957 -11.0648 -10.567

    Regards,

    Andy

  • 0 in reply to Andy Yang78
    TI__Guru 56215 points

    Hi,

    The first row looks correct to me, however, I have not had time to validate the result from other row. If you can make the data as 1 row I and 1 row Q, then it will be easier for me to validate he data.

    The equation you are using is correct. In which file did you produce the data you posted originally? From the above comment, the first 3 rows of Pab is pretty much within a certain range. Did not spread a lot. 

     

  • 0 in reply to Andy Yang78
    TI__Guru 56215 points

    Hi,

    Since I have not heard back from you for a while, I am closing this thread due to inactivity.