▲ 92 points
back
41 comments
> affordable for students and RF enthusiasts
Takes a lot of scrolling to see the price, which is $480. That's probably affordable considering the specs. But the cheap way to experiment with an SDR is to repurpose a $20 USB TV tuner.
As long as you don't plan to transmit. A 20 bucks tuner is cute, but won't get you too far, especially with noise and heat problems affecting signal quality. But yeah, a generally nice way of getting to know sdr and whether you're interested in it at all before spending serious money on grown up boards.
Even the HackRF clones are a quarter of the cost and are much more capable than a 20-40 dollar SDR
For the frequency ranges in which they overlap the rtlsdr dongles out perform the HackRF. I bought my first rtlsdr in early 2013 and am a HackRF kickstarter backer. The BladeRFs are great if you want to do serious work like running cell basestations. That extra price you pay is worth it in that regard.
The cheapest way is to start with a RTL SDR tuner and for example Gnuradio.
If you want to experiment with celluar mobile radios LTE you need a LimeSDR, BladeRF and a help processor.
"If you want to experiment with celluar mobile radios LTE you need a LimeSDR, BladeRF and a help processor."
Why would you need a LimeSDR and a BladeRF ?
I have done plenty of (rx only) GSM/LTE functions with only the (v1) BladeRF ...
If you really want LTE, it's more about xtrx than LimeSDR and BladeRF.
What is a help processor?
Help CPU I meant, something accelerating the signal processing from the software defined radio often a FPGA, field programmable gate array.
In context of GnuRadio and similar, do any projects already utilize FPGA acceleration as offered by USRP?
From what I've seen, it all happens on the CPU.
Adalm Pluto comes to mind.
HackRF one can be purchased for $100 nowadays:
https://www.aliexpress.com/item/HackRF-One-1MHz-to-6GHz-SDR-...
Is it worth paying $380 more to have a better ADC and a few more bells and whistles in the BladeRF micro?
IMO, the bladeRF is more of an alternative to USRPs, and the HackRF is more of a hacker tool.
Full-duplex
Not sure how impressed I am with the specs. It is nice that you can transmit 54MHz of bandwidth or whatever... but what can you do that's actually interesting? It doesn't go down to HF where you can communicate around the world. It doesn't have the oscillator stability to run WSPR in VHF. What do people that transmit with this actually do that a $20 (or even $300 HT) FM radio wouldn't?
There are ways of achieving lower frequencies with SDRs, for example the older model of BladeRF has an expansion card that goes down to 60Khz.
There are tons of things you could transmit, for example GSM, DVB, ATSC, GPS, amateur digital modes, the list goes on.
Voice/CW aren't really what SDRs are for, just like personal computers aren't merely a replacement for typewriters. If all you want to do is voice/CW, then yes, you should just buy a Kenwood or Yaesu. You can do the same things with an SDR, but that's not the point of them.
Nice to have another MIMO radio in the mix.
When I first started experimenting with SDRs I thought "widest frequency range" was the spec to go for, then it became "widest bandwidth" was the spec to go for, and then it because "highest dynamic range" was the spec to go for.
The truth is, depending on what you're interested in, any or none of those could be on your "must have" list.
My journey went as follows; first I wanted to see as many different signals as possible from as many different sources as I could find. That suggested the wider the frequency range the more success I would have. But those sources transmit with a modulation and that modulation might consume a little (CW) or a lot (OFDM) of bandwidth. So more bandwidth, especially if something was using multiple "channels" in a space became key. But what if you can't hear what it is you are listening for? As digital devices the number of bits in your ADC really impacts your ability to deal with large adjacent signals in the band. You find yourself buying (or building) filters to knock back the US FM band, or a nearby TV station for example. For doing some bluetooth work I ended up with a band pass filter on the front end that killed off nearly anything outside the 2.4Ghz ISM band.
Multiple inputs and outputs became a thing for me because I wanted to build an LTE base station, so I backed the LimeSDR and got one of those. There are interesting things to see when you have multiple receivers that are all running from a phase coherent local oscillator. You get this for 'free' when you have multiple receivers running off the same local oscillator.
Lower frequencies are their own problem, the physics gets in the way. Fortunately it is well understood how to heterodyne DC to 200Mhz up into a range that these SDRs will operate. Of course that teaches you about things like phase noise and how it impacts your ability to separate individual channels.
Lots of interesting things to learn, but it isn't just 'plug and go' with one box that can see "DC to Daylight"[1]. Filters, to keep out the signals you don't want. Amplifiers that bring up the signals you do want, and not too much noise. Bandwidth to capture the full signal and a bit extra. Enough sensitivity to pull out the modulation frequencies uniquely. And a fast enough DSP platform to turn that mush into the actual signal you are looking for. It is quite the pile of projects.
[1] Daylight being visible light RF signals :-) Radios that cover say 1MZ to 25Ghz can easily cost $100,000 so they aren't really the stuff you'll pick up casually, but the industry is investing heavily and what you can do for as little as $100 (ADALM-PLUTO) is pretty amazing.
Yeah, I've interacted with a lot of people that didn't appreciate the problems of ridiculously high (2+ GHz) instaneous bandwidth. It sounds great and is easy to sell, but the fact that a single high power signal can wreck everything else is pretty problematic. At that point your options are: 1. Let the high power signal drive the receiver rail to rail, and which point your signals of interest are lost, 2. Set gain to avoid the highest power signal causing clipping, at which point your signals of interest are again lost, this time below the noise floor, or 3. Like you say, add a bunch of filters for the high power undesirable signals, at which point massive bandwidth starts to look less appealing.
One unfortunate aspect of many of these new popular SDRs is that they do not operate at lower frequencies, so you are going to miss all the HF/MF/LF fun, and part of VHF spectrum. For example the 6 meter amateur radio band is not directly accessible with this SDR.
I wonder what it would take to make a DC to VHF (300ish MHz) SDR to complement these UHF SDRs
The original BladeRF has an optional transverter that gives you 60kHz to 300MHz. (I have one but haven't done anything with it yet.)
https://www.nuand.com/product/hf-vhf-transverter/
I don't see one for the version 2 though.
I guess using upconverters like this one is the easiest way to pick up all those bands when your equipment is limited:
The Micro has the same expansion header as the original BladeRF, so I assume they will release a transverter expansion board for the Micro at some point.
Unfortunately the software support for the BladeRF 2.0 micro is not yet very mature - the official installation instructions do not work (don't install udev rules etc).
I would recommend waiting a couple of months and see how the ecosystem develops.
What board / software is the current leader in this space? Is there a brand akin to Arduino ?
The HackRF ($300) or RTL-SDR ($20, but RX only) are the go-to hardware for consumers/hackers. These devices are more limited than some of the newer SDRs, but the software support is a lot better.
For software, the HackRF is probably the best ecosystem. It's open hardware, and the firmware is GPLv2.
The software ecosystem for SDR is pretty messy, pretty much all of the consumer/hacker SDR software is hard to set up correctly, and Linux only. The easiest thing to use is the Pentoo distro, it has all of the common tools, up to date and included in the base image.
Installs fine here on Linux.
USRP radio was 2000$ plus daughter board cost for specific bands. The price of this is so cheap; you can pop garages, car tire sensors, key fobs, cell phone signals, all in one.
With 61 MSPS sample rate, why they spec their filter to 54 MHz? Wouldn't that cause aliasing?
No it's the other way round; with 61 MSPS sampling rate, the theoretical maximum bandwidth is 61 MHz, so they stay below.
[edit]
If you're missing the factor two, as in "sampling frequency must be twice the maximum signal frequency", the keyword is complex sampling. With complex signals, the sampling theorem is "sampling frequency must be greater than the maximum signal frequency".
[edit2]
I don't want to make anybody read about sampling. So we have a real signal - we're in the real world - and suddenly there's a complex signal? You basically fork the signal, and delay one signal path by half the clock period. You can then sample both signals at the same time, and collect the data you would get when sampling with the double frequency. Because we sample two signals, we store two samples at a time. One we call the real signal, one the imaginary -- and there we got our complex signal.
For anyone wanting to read more, this article about IQ is very good: http://whiteboard.ping.se/SDR/IQ
Thanks for posting that. Very nice.
Are you referring to iq modulation here? So in practice they have two 61 MSPS ADCs? I guess it makes sense then to have twice the bandwidth too.
Yes, IQ modulation / QAM.
I haven't studied the BladeRF layout, I can just tell about the Ettus USRP. There, the receive path looks like that:
[Antenna]---[ADL5380 Quadrature Demod]===[2x low-pass filters in parallel]===[2x ADC in parallel]
Then you have digital data that is fed to the FPGA.
Yeah, the BladeRF uses an AD9361[1] as the transceiver, which features a pair of ADCs per channel, of which there are two (so, a total of 4 ADCs and 4 DACs in the package). It's got a bunch of other convenient features like built-in PLLs both LO and baseband.
[1]
Can this transmit more than 1Ghz of spectrum? I can’t see that from the description
Newbie question: what interesting applications does this technology have?
There's an infinite list of things you could do, for example:
- wireless pentesting of GPS, RFID, NFC, garage doors, car remotes, etc.
- Receive NOAA satellite weather imagery
- Track airplanes using ASD-B
- Amateur radio voice/morse code/digital modes
- Run a GSM base station using OpenBTS
- Capture every FM radio broadcast in your area, simultaneously
- Transmit a DVB or ATSC stream to your TV
You could create your own LTE station
> 47MHz to 6GHz
Ok wow!
By itself, that isn't particularly impressive - all that requires is a decent LO + mixer. Make sure you aren't confusing frequency range with instantaneous bandwidth.