0
\$\begingroup\$

I am trying to implement the transmit section of this GNU Radio transceiver design, documented here.

This design is intended for a bladeRF SDR, but the hardware available to me is the USRP B210.

The final block of the script, which was originally an Osmocom Sink block, was switched for the UHD: USRP Sink block to match the hardware. In this block, I set the sample rate to the tx_rf_sample_rate variable = 8 Msamples/s. I set center frequency to tx_rf_freq variable = 464.938 MHz and bandwidth to 6 MHz.

The original audio input was Audio Source block with a 25 kHz sample rate. Instead, I am attempting to use a .wav music file source with a sample rate of 22.05 kHz.

My test setup:

  • Input is an audio file with a 22.05 kHz sampling rate
  • GMRS wire antenna: enter image description here
  • Transmitter is the USRP B210 mentioned above running the GNU radio script
  • Receiver is a Motorola Talkabout T100 - FRS/GMRS 2-way radio. I listen to the audio through its speaker as I walk around.

Issues: During this test, the range is very short (about 50 ft/15 m) and noisy. If I use another Motorola Talkabout T100 as a transmitter, I can hear speech significantly further away with little noise.

My first thought was a gain issue, so I adjusted the gain in the UHD USRP Sink block across its full range, and while it distorts the signal indicating high gain, range is not improved and the noise persists.

Another guess is that the noise is caused by improperly set filters due to differences in sample rates. The BladeRF datasheet specifies a max sample rate of 8 Msamples/s, but the USRP B210 has a max sample rate of 61.44 Msamples/s.

Any assistance in debugging the noise and range issue would be helpful.

\$\endgroup\$

1 Answer 1

Reset to default
2
\$\begingroup\$

Your antenna requires a ground plane to be effective.

It's a magnetic mount (magmount) antenna, that is to be placed on a metallic plate having a minimum diameter of 300 mm (the larger the better).

The ideal location for a magmount antenna would the top of a car (it was designed for that!).

It wouldn't work without a ground plane and with the coaxial cable coiled up.

The range would improve significantly should the magmount antenna, with the 300 mm diameter ground plane, be mounted at a height.

\$\endgroup\$
4
  • \$\begingroup\$ Thanks, for the response. I grabbed a copper sheet (about 2ft x 3ft) as a ground plane and uncoiled the cable and that definitely made a difference. Also, I ran some tests to measure the output power of the antenna \$\endgroup\$
    – choga12
    Commented Aug 12, 2022 at 18:47
  • \$\begingroup\$ Also, I ran some tests to measure and compare the output power of the magmount antenna and Motorola Talkabout T100 at multiple distances. The first point to note from this test is that I transmitted on FRS Channel 2 (462.5875MHz). I found that the Motorola Talkabout T100 transmitted at 462.5MHz and the GNU Radio design transmitted at 463.38MHz. This is about a 0.88MHz offset from the intended frequency. \$\endgroup\$
    – choga12
    Commented Aug 12, 2022 at 19:00
  • \$\begingroup\$ Another point is that the Motorola Talkabout T100 had a significantly higher power level throughout each distance compared to the magmount antenna. For reference my noise level was around -70dBm. Starting with the Motorola radio, the 4 test distances measured power levels at -9.6, -17.46, -18.93, and -31.08 (dBm), while the magmount antenna at the same 4 test distances had power levels at -59.2, -52, -54, and -69 (dBm). Would you have any ideas why the frequency had shifted when using the GNU Radio design? Or do you have any ideas to increase the power level of the magmount antenna? \$\endgroup\$
    – choga12
    Commented Aug 12, 2022 at 19:10
  • \$\begingroup\$ Hi choga12, Good that the ground plane and the cable did make a difference. Yes, the difference in frequency steps between the two radios could also cause noisy reception. I'm sorry I'm not conversant with the GNU radio design. \$\endgroup\$
    – vu2nan
    Commented Aug 13, 2022 at 2:52

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.