Cte data caculate arrival angle

Hi,

     I am using nrf52833 for AOA, the examples are direction_finding_connectionless_tx and direction_finding_connectionless_rx. Now I use two antenna array for locater and I can get cte data. For further caculating I print I and Q data in cte_recv_cb function. My question is no matter how I adjust the arrival angle, the caculated results are barely change(all results are about 90 degree), while the I and Q data change a lot.

    The distance between two antennas is 12cm. I caculate phase angle of  each sample data use arctan(Q/I) (use atan2 in python can get  radian value). Then from the 9th sample value, caculate the difference in pairs. Use cofficient 12.5/(2*pi) to get the cosine of angle of arrival.

    In this way of calculating, the difference always less than 10 degree, and caculated results close to 90 degree no matter how I change angle of arrival. Could you please help me to see if there is any problem with the calculation method. Here are some I Q data

I : [-31, 30, -30, 31, -33, 37, -33, 32, -38, -41, -39, -45, -40, -46, -40, -45, -37, -42, -35, -37, -29, -35, -22, -29, -18, -20, -11, -17, -3, -6, 6, 1, 11, 9, 20, 14, 26, 24, 30, 33, 37, 38, 38, 41, 38]
Q: [33, -33, 34, -32, 31, -28, 31, -29, 13, 18, 8, 12, -5, 2, -8, -8, -17, -13, -20, -24, -27, -28, -34, -35, -36, -41, -41, -42, -40, -44, -39, -46, -39, -46, -36, -43, -30, -37, -29, -31, -19, -28, -12, -15, -5]

I: [-34, 36, -37, 35, -37, 34, -35, 37, -63, -40, -66, -43, -65, -45, -68, -45, -61, -42, -61, -39, -48, -34, -42, -26, -32, -19, -21, -10, -12, -9, 5, -2, 17, 9, 31, 19, 36, 27, 46, 31,
53, 37, 59, 41, 63]
Q: [26, -27, 29, -27, 27, -29, 27, -27, 19, 18, 11, 12, -5, 4, -12, -5, -25, -15, -27, -25, -44, -30, -51, -35, -58, -42, -63, -43, -66, -44, -67, -44, -64, -44, -60, -41, -54, -37, -46,
-32, -39, -23, -30, -21, -19]

Parents
  • Hi, 

    The raw IQ data that is transmitted must be converted into vector form to be reviewed properly. We don't have any specific sample code showing this, but you can check out this page for information on the various formulas used to translate IQ data between polar and rectangular forms. Nordic will not provide a Direction Finding algorithm for you I'm afraid. 

    It could be a problem with your Antenna array also for all I know, as you don't say what kind of antenna array you're using. Does "two antenna array" mean that you only have 2 antennas in your array for example? If so that is too few antennas. They will need a minimum of 4 antennas (2x2 array) to get usable IQ data. 

    Regards,
    Amanda H.

Reply
  • Hi, 

    The raw IQ data that is transmitted must be converted into vector form to be reviewed properly. We don't have any specific sample code showing this, but you can check out this page for information on the various formulas used to translate IQ data between polar and rectangular forms. Nordic will not provide a Direction Finding algorithm for you I'm afraid. 

    It could be a problem with your Antenna array also for all I know, as you don't say what kind of antenna array you're using. Does "two antenna array" mean that you only have 2 antennas in your array for example? If so that is too few antennas. They will need a minimum of 4 antennas (2x2 array) to get usable IQ data. 

    Regards,
    Amanda H.

Children
No Data
Related