Menu

Transmit-receive experimens: Is SNR of 20dB good enough?

Reception
2018-08-08
2019-02-07
  • Joel Lehtonen

    Joel Lehtonen - 2018-08-08

    I'm doing "closed circuit" transmit-receive tests with two SDRs, RF attenuators and Dream in both transmit and receive modes on different hardware.

    At transmitter end I'm using LimeSDR Mini and Dream in transmit mode with the following modulation parameters:

    interleaver=SI_LONG
    msc=CS_3_SM
    protlevel=1
    robustness=RM_ROBUSTNESS_MODE_A
    sdc=CS_2_SM
    spectocc=SO_4
    

    At receiver end I'm using RTL-SDR dongle outputting I/Q samples to Pulseaudio null sink and Dream with I/Q (pos) input.

    I found that the minimum signal-to-noise ratio I am able to receive the broadcast is about 20 dB. What is the practically miminum SNR for DRM reception? Is 20dB a good result or should I look for any error sources from my SDR pipeline?

     
  • Julian Cable

    Julian Cable - 2018-11-09

    I'll pass your question on to someone who can offer a better answer than I can. You might also find some useful information in the broadcasters guide and the mimimum receiver requirements here.

    There are a few people playing with the transmitter so I've added a new forum. If you post on there you might get a little community together.

     

    Last edit: Julian Cable 2018-11-09
  • Simone Stoeppler

    Hi Joel, 20 dB will not be good enough to decode all transmissions, depending on the modulation parameters. In the lab you should at least be able t get 28- 30 dB, with a good path on a real transmission at least 24- 26 dB SNR. You can check some results on the DRMRx forum, i.e. this thread: http://www.drmrx.org/forum/showthread.php?t=1143&page=105&highlight=1296 which was mode A with 26 kbps.

     
  • Stefano Mollo

    Stefano Mollo - 2018-11-10

    Hi, I am also experiennting with TX mode, but with a twist. I am transmitting DRM over IP, and I get SNR of about 65 dB and, needless to say, EXCELLENT full stereo quality on the receiver's end using OPUS as encoder .
    I came to the conclusion that it doesn't matter what physical media is used; as long as it is linear enough, it will work. Try http://172.104.78.117:8000/drm.ogg and pipe the streamed audio via a VAC to DReaM input; L+R / sample rate 48Khz. You'll be amazed.
    I have also experimented with small tiny 433 Mhz modules. It works :-)

     

    Last edit: Stefano Mollo 2018-12-04
  • DL3GBA

    DL3GBA - 2018-11-22

    on this page http://de.intervalsignals.org/kurzwelle-drm/umbau.html you may click on one of the mentioned receivers - on those subpages you may find SNR vs. RF-Level measurements in the Lab.
    Although the pages are in german the diagrams may be helpful (otherwise you may use https://www.deepl.com/translator).

    The result: the determining factor is the linearity in the signal path. This can be seen, for example, in the ripple of the SNR curve on the AOR7030, which is caused by the AGC stages, and the drop of SNR in high RF levels on both the FRG7700 and the NRD525 due to compression and clipping.

     

    Last edit: DL3GBA 2018-11-22
  • Ian Brooks

    Ian Brooks - 2019-02-07

    I have just installed dream_bionic_svn1021 on my 8-year old Acer Laptop and installed pulseaudio Virtual Sink in Ubuntu Mate 18.04LTS. I am using this with an Airspy Mini dongle and a Miniwhip antenna ( RA0SMS design) mounted 5 metres above my garden shed. I have been listening to Radio Kuwait's DRM transmission on 15110 kHz @ 1300GMT today. S/N was around 20dB most of the time and decoding was near perfect. My location is 12 miles north of Bournemouth in Southern England.

     

Log in to post a comment.