4
$\begingroup$

I am wondering if SDRs (software defined radios), e.g. HackRF, RTL-SDR, or USRPs, are, in essence, just oscilloscopes you can plug to a computer to get the data from it. If not, what are the main differences between such devices ?

$\endgroup$
1
  • $\begingroup$ RTL-SDR was a USB based terrestrial TV receiver for PCs... with the advent of internet based streaming and smart phones, the blue-sticks lost their jobs... Eventually their dlls are hacked for good reasons to convert them into very flexible software defined radio receivers... :-) oscilloscope ?? $\endgroup$
    – Fat32
    Nov 22 '21 at 20:44
10
$\begingroup$

All the three devices you mention have mixers, so that's what you wouldn't find in an oscilloscope usually. The amplifiers are optimized for weaker signals, and typically, the quality of ADCs in expensive oscilloscopes is higher than in SDRs – because if you're limited by noise and interference, a high-resolution ADC has very limited benefit.

Also, (higher-end) oscilloscopes are (often) optimized for very high sampling rates streamed into finite-sized memory for display, whereas SDRs are meant to stream continuously.

An SDR is a radio device, an oscilloscope a measurement device.

You can build one out of certain members of the of the other class – the technical difference between the acquisition stage in an oscilloscope and a direct-sampling SDR might not be that large on paper – but clearly, the things that make these devices what they are are not just the ADC!

$\endgroup$
6
  • 2
    $\begingroup$ Also: oscilloscopes are designed to have flat and well-behaved frequency response, while especially cheaper SDRs can have widely varying sensitivity in different bands. $\endgroup$
    – jpa
    Nov 23 '21 at 7:42
  • $\begingroup$ Thanks for the answer. Small question, when you say "the amplifiers", you mean the amplifiers in oscilloscopes ? It might be another question, but why would oscilloscopes not use mixers in the same way as SDRs ? $\endgroup$
    – R. G.
    Nov 25 '21 at 14:53
  • 1
    $\begingroup$ @R.G. both scopes and SDRs have amplifiers, but they're for different input powers, and have different noise figures. $\endgroup$ Nov 25 '21 at 16:27
  • 1
    $\begingroup$ @R.G. in which case would you want your scope to have a mixer? That only makes sense if the signal you're observing is modulated to carrier higher than its bandwidth, which for oscilloscope applications is usually not the case. $\endgroup$ Nov 25 '21 at 16:27
  • $\begingroup$ So an oscilloscope is usually used for signals from 0 Hz to a certain frequency, while SDRs are used for only a small bandwidth which is not "centered" on the DC, am I right ? Thanks for the precision. $\endgroup$
    – R. G.
    Nov 25 '21 at 18:24
3
$\begingroup$

That depends on how much your definition of "just an oscilloscope" limits functionality. There's probably enough hardware in a high-end O-scope to make at least a receive-side SDR.

The big difference: An oscilloscope lets you see a waveform. A basic oscilloscope just needs to collect a series of samples and display a graph. A basic oscilloscope also gathers information at the same rate that it presents it, but might narrow down the information saved by only acquiring a chuck of it in time.

On the other hand, a radio takes RF, and processes it into something informative (data, voice, pictures, whatever), or it takes something informative and turns it into RF. A radio receiver generally has intermediate steps where it narrows down the frequency spectrum of the received content, but does not take chunks in time.

$\endgroup$
1
  • $\begingroup$ Thanks @DanBoschen: changes made. $\endgroup$
    – TimWescott
    Nov 23 '21 at 0:37
0
$\begingroup$

If you have an rf front end that can do the down conversion, yes.

$\endgroup$
6
  • 1
    $\begingroup$ An RF downconversion in front is the scope isn’t necessary per se- an LNA and bandpass filter will be sufficient for much of the occupied radio spectrum with today’s high sampling rate oscilloscopes. For instance I have demonstrated the reception of GPS signals using a Lecroy sampling scope (15 years ago!) with the scope set to a 20 MHz sampling rate. No external down-converter was necessary- just the LNA and filter! $\endgroup$ Nov 22 '21 at 22:00
  • $\begingroup$ I'm assuming the scope can't sample fast enough for direct conversion. Obviously in a perfect world you just use a LNA and dsp. $\endgroup$ Nov 23 '21 at 0:02
  • 1
    $\begingroup$ By the way, why do you think SDR are still so expensive? I can get a gig/s sampling scope for 350 on Amazon but in an sdr it would be exponentially more expensive. $\endgroup$ Nov 23 '21 at 0:41
  • 1
    $\begingroup$ That would be a good question to post if you can match up specs that you believe are equivalent! $\endgroup$ Nov 23 '21 at 1:02
  • 1
    $\begingroup$ @FourierFlux Usually SDR can process samples continuously and a scope only for given waveform length. $\endgroup$
    – jpa
    Nov 23 '21 at 10:48
0
$\begingroup$

The biggest difference is that SDR is more of a spectrum analyzer where multiple transmitters can be decoded simultaneously. Scopes are for time domain whereas spectrum analyzers are for the frequency domain.

Also SDR has band switched input filters to minimize front end overloading and getting a much better SNR for the weaker signals of interest, you just cant do that with a scope unless it has a spectrum package built in (in which case its a spectrum analyzer!).

I use SDR for pre-testing in EMC work but I would never use a scope connected to an antenna, that would be silly.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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