Combining the bandwidth of 2 RTL-SDR dongles in GRC

A while ago, I had the plan of taking 2++ RTL2832U SDR dongles to get a wider spectrum in a FFT-plot. Since I had no idea of gnuradio-companion then, I didn’t know how to do this. But thanks to my purchase of a HackRF One, I was forced to use GRC and found some very useful tutorials about SDR in general, with special attention to the HackRF One. You can find the lessons by Michael Ossmann here: http://greatscottgadgets.com/sdr/

So, now that I knew a thing or two about gnuradio, I opened an empty flowgraph and this is the final outcome with 4.4 MHz of bandwidth:

multi_dongle_grc

4.4_rtlsdr

I simply took two RTL-SDR dongles at their max. bandwidth of 2.4 MHz, resampled the signals to 4.8 MHz, then shifted the first signal down by 1MHz, the other one 1 MHz up, added them together, divided the combined signal by 2 to lower the noise floor again and finally feed it into a FFT plot.

At first, I tried shifting the signals by 1.2 MHz to get full 4.8 MHz , but I realized, that I had a notch in the center, so I reduced the frequency shift until I had no notch anymore.

4.8_rtlsdr

Compared to the HackRF One at 4.8 MHz sampling rate:

4.8_hackrf

Notice, that this isn’t any useful to decode or demodulate any signals, because of phase and clock misalignment, but it can be very handy, if you want to view a larger spectrum than 2.4 MHz and don’t want to spend money in a more expensive SDR, that can be used at higher sampling rates.

If you want to check out the flowgraph or edit it for even more RTL-SDRs, feel free to download it here:

4.4 MHz GRC flowgraph

4.8 MHz GRC flowgraph (with notch at center frequency)


Comments

  1. Airspy? Any one?

  2. This is really neat. I’ve been wanting something like this for years but didn’t realize it could be this “simple” when you don’t need phase coherence and all that jazz. Thanks!

  3. Milan van de Swaluw
    8. January 2016 - 10:44

    You could always give two RTL-SDRs the same phase and clock by adding an external clock as shown in: http://pe1ryy.blogspot.com/2015/04/nt7ss-si5351a-vco-hooked-up-to-local.html && http://pe1ryy.blogspot.com/2015/04/low-pass-filter-added.html

  4. the signals won’t be phase coherent. the dongles would only operate with the same clock

  5. With the multi rtl block from Piotr Krysik you can presync the two dongles on a given freq. I was able to decode without prob a gsm signal with center freq at the middle of the two combined bandwidth with your grc. If you want my grc let me know ;) Note that this multi rtl block is not very stable at the moment and stop running after few minutes

  6. Syed Ghazanfar Ali Shah Bukhari
    7. November 2016 - 23:02

    Dear Oliver
    I have used your grc flowgraph and modified it to be used with 2 HackRF Ones I also shifted the two bandwidths inward by 1 MHz instead of 0.2MHz to make a smooth joint for a 38 MHz spectrum. Unfortunately one of my HackRF Ones has its RF Amp burnt up so I adjusted its IF and BB gain to have same noise floor as that of other HackRF One. Its really awesome. I have sent the diagram to rtl-sdr blog having your name as such and your website link.
    Thanks & Stay Blessed

  7. Coherent Receiver
    14. November 2016 - 11:15

    Just for all, who would like to make experiments. We had developed a multi-channel coherent receiver that enables to aggregate N-RTL-devices (http://coherent-receiver.com). You can test the bandwidth extension or coherent averaging.

Leave a Reply

Your email address will not be published / Required fields are marked *