Forgot your password?
typodupeerror
Communications Input Devices

Software-Defined Radio For $11 171

Posted by timothy
from the us-dollars-of-course dept.
Malvineous writes "Don't have $1500 to drop on a USRP? A Linux kernel developer has discovered that a Realtek digital TV tuner chip has an undocumented mode that turns it into a software-defined radio, with a frequency range of 64-1700MHz. The going rate for one of these USB devices can be as low as US$11. If you're unfamiliar with software-defined radio and have 20 minutes to spare, Balint Seeber has a great video introduction."
This discussion has been archived. No new comments can be posted.

Software-Defined Radio For $11

Comments Filter:
  • /.'d - google cache (Score:2, Informative)

    by Anonymous Coward on Saturday March 31, 2012 @03:29PM (#39535919)
  • by Nyder (754090) on Saturday March 31, 2012 @03:37PM (#39535975) Journal

    rtl-sdr

    DVB-T sticks based on the Realtek RTL2832U can be used as a cheap SDR, since the chip allows transferring the raw I/Q samples to the host, which is officially used for DAB/DAB+/FM demodulation. The possibility of this has been discovered by the V4L/DVB kernel developer Antti Palosaari.
    Specifications

    The RTL2832U outputs 8-bit I/Q-samples, and the highest theoretically possible sample-rate is 3.2 MS/s, however, the highest sample-rate without lost samples that has been tested so far is 2.8 MS/s. The frequency range is highly dependent of the used tuner, sticks that use the Elonics E4000 offer the best range (64 - 1700 MHz).
    Supported Hardware

    So far, the following devices are supported:

            ezcap EzTV668 USB 2.0 DVB-T/DAB/FM stick (Elonics E4000 tuner) (sources: AliExpress, Dealextreme)
            ezcap EzTV666 USB 2.0 DVB-T/DAB/FM stick (Elonics E4000 tuner, picture Download)
            Hama nano DVB-T stick (Elonics E4000 tuner)
            Terratec NOXON DAB/DAB+ USB-Stick (Fitipower FC0013 tuner)

    People over at reddit are collecting a list of other devices that are compatible.

    Other sticks based on the RTL2832U might be added in the future as well.

  • Re:Too long (Score:5, Informative)

    by shish (588640) on Saturday March 31, 2012 @03:38PM (#39535987) Homepage
    Watching the video now, it appears to be "dump ALL the radio signals to disk, for later analysis", so you can then use software to pick out audio / text / morse code / etc signals
  • Re:Too long (Score:2, Informative)

    by trout007 (975317) on Saturday March 31, 2012 @03:39PM (#39535991)

    I found this great new site you may be unaware of. It's called Wikipedia. It is kinda of like an online encyclopedia that has brief summaries of almost anything. Check it out I'm sure you will like it.

    http://en.wikipedia.org/wiki/Software-defined_radio [wikipedia.org]

  • Re:Too long (Score:3, Informative)

    by MinusOne (4145) on Saturday March 31, 2012 @03:47PM (#39536049)
    Would it really have been that hard to embed the Wikipedia link in the article? Sure I can look this some up, but someone is trying to explain it to me and its just one stupid link.
  • Re:Too long (Score:4, Informative)

    by Polo (30659) * on Saturday March 31, 2012 @04:01PM (#39536151) Homepage

    I found this particular video showed me what you could do in a visual way I wouldn't have picked up by reading about it.

    Yes, I agree that sometimes you can't watch a video where you could read text though.

  • Challenge Accepted (Score:5, Informative)

    by Gordonjcp (186804) on Saturday March 31, 2012 @04:30PM (#39536315) Homepage

    In a conventional radio receiver, you start by filtering off the wanted signal with a broad filter, mixing it with another locally-generated signal (the Local Oscillator) to make a lower Intermediate Frequency (IF), then filtering the IF to extract a single "channel" of information. Then you demodulate this, possibly after mixing it down to an even lower IF.

    In a software-defined radio, you convert directly down to a much lower frequency (audio frequency, even), but - and this is the clever bit - you do it with two local oscillators, 90 degrees out of phase. This gives you a complex sample, a pair of samples representing In-phase and Quadrature, or the real and imaginary components of your signal.

    From there you can apply digital signal processing techniques to extract the wanted signal, show an FFT of the chunk of band you're capturing, and so on. This lets you do very sharp filtering, because you're no longer constrained by the physical realities of trying to implement electronic filters with practical components.

    Shameless plug - if you want to try SDR out, go here:
    https://github.com/gordonjcp/lysdr [github.com]
    Follow the instructions in the README, then either post a reply or bug me in irc.freenode.net ##electronics for further instructions.

  • by ctrl-alt-canc (977108) on Saturday March 31, 2012 @04:46PM (#39536429)
    A few years ago, together with a friend, we reverse engineered a DVB-T usb pen by Hauppauge, and we were able to extract the raw data stream skipping the demodulation process. We did it since we wanted to test if it the device could be used as a DSP IF strip in a homemade spectrum analyzer. The device worked, but the analog IF strip we wanted to replace was actually drawing circles around its digital replacement, so we abandoned the project. 8 bit of resolution and an hardware designed for a very specific purpose couldn't bring us too far, as we feared.
    It is nice to see that somebody else was capable to reverse engineer these devices, but as you can see from their results, they aren't actually that good. I saw somewhere that a USB pen for DVB had to hit the market, and its ADC has been announced to be 12 bit wide.This could be an interesting device to hack for SDR applications, hoping it isn't vaporware...
  • Re:Too long (Score:4, Informative)

    by Grieviant (1598761) * on Saturday March 31, 2012 @05:48PM (#39536807)
    The modern definition of 'radio' refers to any device that uses wireless to transmit / receive an electromagnetic signal. Your wifi modem, bluetooth device, smartphone, TV, etc., are all radios. It's not limited to traditional FM radio from 88-108 MHz. Pretty much all radios are using some form of digital communications these days, as opposed to analog modulation with FM radio. Software defined radio really isn't anything special - the low frequency part of the radio (from IF down to baseband) is handled with a digital signal processor that can be programmed to handle many different formats. It's a natural evolution of digital communications that gets tossed around as a buzzword.
  • by Anonymous Coward on Saturday March 31, 2012 @06:17PM (#39536953)

    You realize that if a significantly-sized company is selling you a radio for $1K that has $150 worth of parts in it, for an item that isn't mass-market consumer
        (and ICOM/Kenwood/Yaesu radios intended for the HAM market aren't "mass market", compared to, for example, cell phones), that they are on the edge of
        *losing money*. I've run two different manufacturing businesses in the last 20 years, and really, if you only look at the raw BOM costs of something, and don't
        factor in all the other costs of bringing a reliable product to market, you'll go out of business.

    Many of the so-called "high dollar" SDRs out there are "high dollar" for fairly good reasons. They are typically sold into markets where economies of manufacturing scale don't really apply, they often have feature sets that are *vastly* larger than what we're seeing in these "rtl-sdr" devices, and they tend to use higher-quality components. These DVB-T devices, for example, use a master clock that is good to about 100PPM--which for radios is rather seriously crappy. They won't have features like a DDC (usually FPGA resident) for fine-tuning the RX signal. The 8-bit resolution may be fine for some applications, but for others requiring higher dynamic range, that 8-bit resolution will be a killer. Plus many of the "high dollar" SDR devices offer TX chains as well as RX chains, and all the other comments apply for the TX chain as well.

    But one of the big things about "high dollar" SDR devices is that they're primarily designed as *development platforms* for developing SDR applications across a wide "spectrum" of fields of endeavour. So they include large FPGAs, those FPGAs allow you to perform part/all of your DSP algorithms at insane speeds inside the FPGA--speeds/sample-rates that would be impractical for a host-software implementation. Large/fast FPGAs are expensive, and that cost has to be passed on. Further, the "high dollar" SDRs typically offer bandwidths into/out-of the host at much higher sample rates than 3.2Msps. Yes, 3.2Msps, RX-only, 8-bit resolution, no-fancy features is entirely-adequate for a lot of different hobbyist work. But it's inadequate for a lot of other types of work for which the so-called "high dollar" SDRs are supremely-well suited.

  • by Anonymous Coward on Saturday March 31, 2012 @06:59PM (#39537195)

    Jesus tapdancing christ, a finished product is more than the sum of its parts - you sound like the people who whine because Apple's making $x "profit" based on $sale_price - $parts_price.

    1) Much higher quality hardware filters for the band and for the particular AF signal required, giving the sort of selectivity needed for amateur levels of power;
    2) Low noise amps, again essential for amateur levels;
    3) DDS or PLL tuners accurate to a few Hz at worst, not ones which vary by at least 100Hz on the tuned frequency;
    4) Well-documented open or semi-open RS232/USB interface;
    5) Higher resolution ADC;
    6) DSP running on-board rather than wasting the CPU cycles of your far hotter desktop/laptop;
    7) Good, local vendor support for a small market keen to get the most use out of a well-built product, rather than zero vendor support for a consumer market buying a throwaway item;
    8) Sometimes an on-board UI.

    Finally, you're ignoring that "Icom etc" make transceivers, whereas this is firmly receiver-only. And a shitty receiver is one thing, but someone operating a transmitter with shitty parts isn't only being a dick to fellow spectrum users but may be breaking the law.

  • eBay (Score:4, Informative)

    by ArchieBunker (132337) on Saturday March 31, 2012 @07:06PM (#39537239) Homepage

    But the sellers have sold out days ago along with all the other retailers. Looks like one is left and they ship direct from Hong Kong so expect a few weeks wait. I'm kinda pissed about only finding out now.

    I've been toying with the idea of getting an entry level SDR for a while just to see what all is out there on the airwaves. This USB tuner chip appears to do the same thing as the $100+ Funcube Dongle.

  • by Anonymous Coward on Saturday March 31, 2012 @07:46PM (#39537479)

    AMPRnet has been around since the late '70s, various forms of packet radio existing before and after. The first TCP/IP stack I ever used was KA9Q.

    1) lack of bandwidth;
    2) legality of high power transmissions for non-hams;
    3) legality of encrypted transmissions for hams.

  • by colsandurz45 (1314477) on Saturday March 31, 2012 @09:34PM (#39538041)

    All recent RF chips for wireless are zero IF nowadays.

    Not true, super heterodyne is still very popular.

    A SDR approach will use a big DSP (vector DSP even) to do the processing in software.

    Not really. Depending on your platform the industry trend is actually going away from DSPs. DSP operations are being implemented in FPGAs these days (since they're faster and the newer Virtex 6 or 7s (and whatever Stratix whatever) are really huge).

  • Re:Too long (Score:4, Informative)

    by wagnerrp (1305589) on Sunday April 01, 2012 @12:06AM (#39538619)
    I'm sure 99.999% of the population who thinks "radios" are something they're listening to music on have exactly zero reason to know what a software defined radio is. The rest remainder of us who actually understand technology can use whatever terminology we see fit.
  • by Guy Harris (3803) <guy@alum.mit.edu> on Sunday April 01, 2012 @02:47AM (#39539075)

    "I found this great new site you may be unaware of. It's called Wikipedia. It is kinda of like an online encyclopedia that has brief summaries of almost anything", trout007 Is this the same Wikipedia that says Windows NT [wikipedia.org] wasn't designed for the Internet?

    No, it's the same Wikipedia that says [wikipedia.org]

    Consumer versions of Windows were originally designed for ease-of-use on a single-user PC without a network connection, and did not have security features built in from the outset. However, Windows NT and its successors are designed for security (including on a network) and multi-user PCs, but were not initially designed with Internet security in mind as much, since, when it was first developed in the early 1990s, Internet use was less prevalent.

    which is not the same as "Windows NT wasn't designed for the Internet".

To err is human -- to blame it on a computer is even more so.

Working...