1

I am using an Rx antenna array (4 antennas), and a TX transmitter. Without changing my location of both transmitter and receiver as well as modifying my array geometry, I am measuring different phase shift from each measurement. I am expecting a constant phase shift between each antenna at each trail but it is not the case. Does anyone know why is this?

Detail set-up:

Tx: 2.45GHz carrier at 3.5 meters from the Rx array. Sending a 5kHz pure-tone sine wave at the 2.45GHz carrier. Tx is placed (roughly) orthogonal towards Rx.

Rx: a 2X4 antenna array, both horizontal and vertical distance between adjacent antenna is 10cm. Out of 8, 4 antennas was used in my measurement. Rx array is built onto a flexible substrate but it was staying still for the entire experiment.

Equipment used: Tx--> USRPX310 with SBX daughter board, Rx --> USRPX310 with twinRx daughter board. Tx and Rx are synchronized using Octo-clk for both Ref in and PPS.

Here is an example of plotted based-band waveform:

trail 1, time domain waveform for 4 antennas: enter image description here

trail 2, time domain waveform for the same 4 antennas: enter image description here

The test is done indoor I wonder if the environment would induce varying channel (by multi-path) ? or does no one had this problem before ?(in that case I would check my USRP).

Xingda Chen
  • 11
  • 2
  • 1
    What happens if you connect Tx directly to RX via an attenuator, cable and splitter? That should be your first test, to see whether the effect is in the USRP or the antennas and propagation path. – tomnexus Nov 09 '22 at 01:10
  • Also what does *a 5kHz pure-tone sine wave at base-band* mean - are you transmitting a single tone at 5 kHz, at 2450005 kHz or have you amplitude or frequency modulated the 2.45 GHz signal? What is the scale of x and y axes of the graphs? – tomnexus Nov 09 '22 at 01:26
  • your time domain waveforms seem to be very broken. Software bug on your side? they most definitely look clipped. You're not making sure the envelope of the complex baseband (thus, the magnitude of the complex values) stays within 1. Fix that. – Marcus Müller Nov 09 '22 at 17:00

0 Answers0