Hacker News new | past | comments | ask | show | jobs | submit login
Universal Radio Hacker: Investigate Wireless Protocols Like a Boss (github.com/jopohl)
244 points by mindcrime on Feb 28, 2021 | hide | past | favorite | 44 comments



This made it really easy to figure out how to interpret the packets from my wireless barbeque thermometer. Previously i used some monstrous excel for this kind of pattern finding, e.g. https://github.com/cleanflight/cleanflight/issues/1125#issue...

In the end the bbq temp sensor turned out to be trivial, using rtl433 on a raspberry pi I just log the packets to a dynamo table and alexa (using google charts from a lambda) plots the last 20 points on screen.


Nice hardware is the HackRF One [1], since you can record and re-transmit the recorded signal. Compatible with GNU Radio, SDR#, and more, 1 MHz to 6 GHz operating frequency. [1] https://greatscottgadgets.com/hackrf/one/


Does having a 8-bit converter seriously affect in any way reception?

I'm happy with my rtl-sdr v3 dongle, although living in a very noisy area would like to get something better for HF reception. I don't have a HAM license and no need for the HackRF transmit capabilities, though it may turn out useful for testing remotes, filters, aligning RF stages in diy receivers etc, but the SDRPlay RSPdx or the AirSpy + Spyverter seem better alternatives for receiving only with their 14 and 12 bit ADCs.

Here's a comparison of the three; I'd be interested in reading from other owners as well. https://www.rtl-sdr.com/review-airspy-vs-sdrplay-rsp-vs-hack...


I have a bunch of RTL-SDRs, the HackRF, and the AirSpy. I have only done VHF -> UHF work with them, never HF. In my experience, the HackRF receive frontend is pretty poor, at least in comparison. I'd use it if I needed transmit or for local strong signals only. The AirSpy has the best receiver of the bunch, but worse software support. I've had quite a challenging time getting it to work with GNU Radio and GQRX. It works great in SDR Sharp but that is no longer FOSS and must be run on Windows IIRC. The RTL-SDR is cheap and has a receiver at least on par with the HackRF and by far the best compatibility.


Do you have any experience with the limesdr? I've been eyeing those for a bit


I have a limesdr and a hackrf. Both work really well, but it does seem like there are more people/projects for the hackrf. Once you learn the ins and outs of GNUradio, it is pretty amazing what you can do.


I bought one several months ago, albeit haven’t used it much yet. My one concern is how hot it ran when I first plugged it in, you may want to get a case for it.


I purchased the aluminum case along with my LimeSDR. A few months ago, I figured out how to use trunk-recorder, and the Lime has been my police/EMS radio scanner ever since. It's been running almost nonstop for months, and I haven't noticed any issues -- even as it remained hot to the touch for all those months.


Oh yea, it does run pretty hot! I made a case for it from an electronics project case I had laying around, and even with that it is toasty. If I was building something that was going to be running of for a long time a 3d printed case with support for a small fan would be helpful.


Wow, I was looking at the Blade and Blade mini but this is cheaper.


> 1 MHz to 6 GHz operating frequency.

But what is the maximum bandwidth?


> URH allows easy demodulation of signals combined with an automatic detection of modulation parameters making it a breeze to identify the bits and bytes that fly over the air

Something I’ve been wanting to do for a long while is to figure out how to decode the time code audio signal of Serato Control Vinyl.

I briefly looked at the audio signal of Serato Control Vinyl in Audacity and you can tell that it should be possible to reverse engineer what’s going on. The signal if I remember correctly looks like a blend between a sine wave and a triangle wave, with a sort of “bend” that it looks like half of a banana kind of. Furthermore from memory the frequency of the signal changes over time, the shape changes, and there is a (possibly changing) phase difference between the left and right channels.

Somehow, this encodes a time code. And I can tell that it makes sense that you’d be able to tell the time from looking at the changes in the signal. But I don’t know how to actually decode it in order to determine the time code.

Serato Control Vinyl is used for Digital Vinyl Systems (DVS) in DJing. It allows you to use the control vinyl on a vinyl player and hook it up to your computer in order to control the playback of audio on the computer.

This way you can use physical control vinyl to do scratching and other DJing, and the software on the computer is able to very accurately manipulate the playback of digital audio based on the analog time code signal, so that speed and direction and time position is accurately tracked with high fidelity.

For anyone else who also thinks this sounds interesting, download the control CD from https://serato.com/controlcd/downloads and look at the audio signal, it is similar (possibly even the same?) as the one on the control vinyl.

In fact there is already some third-party software that supports using the Serato control vinyl. For example, the open source Mixxx. https://manual.mixxx.org/2.3/en/chapters/vinyl_control.html

However, Mixxx is GPL v2 licensed, and I am hoping to implement interoperability with Serato control vinyl in software that is either ISC licensed or proprietary, so I don’t want to even look at GPL licensed source code.

If anyone is able to figure out how the time code signal of Serato control vinyl works, or knows of an existing description of it that can be used for implementing software for it, please let me know.


Happy to see another DJ nerd here.

Mixxx's timecode engine is powered by xwax:

https://sourceforge.net/p/xwax/mailman/xwax-devel/ I'm pretty sure there was a thread on the mailing list where they talked about it.

My understanding is that there are "directional" and "positional" signals on the timecode to give you "relative" and "absolute" positioning.

You may be able to glean more from mspinkys as well: http://www.mspinky.com/SoftwareSpec_0.0.6.pdf (no code here, just function names and descriptions)

Good luck!


Thank you :)


That brings back nice memories, I designed some of the early Serato scratch hardware. I didn't work for them directly though, so never did manage to find out how the vinyl timecode worked.


Cool! If I ever run across you IRL I’ve got to buy you a beer and pick your brain :D


What would be a very good starter software defined radio?


Hands down the PlutoSDR (a.k.a ADALM-PLUTO) has the best price/performance & support I've seen (and I've owned a BladeRF, LimeSDR and various RTL dongles as well).

The core radio chip itself (the AD9363) appears to be identical and/or a binned version of the AD9364 chip used in things like USRP Radios (i.e. the high-end standard bearer brand for SDR) and can be configurated to behave like it [1]. The unit price for the AD9364 chip (in low volumes) is around $200. The ENTIRE PlutoSDR is $150 and that includes a Zynq-7010 FPGA (which itself is not cheap)!

Because it's built by Analog Devices, it's generally well supported. It's designed as an educational tool so there are a lot of tutorials for beginners like PySDR. I really like the interfacing libraries (pyadi) that allow me to easily configure the SDR from a python notebook and pull an RX buffer into numpy and process away.

[1] https://wiki.analog.com/university/tools/pluto/users/customi... [2] https://pysdr.org/index.html


Your link says only early builds had the AD9364, and production builds have the AD9363. Too bad, its certainly less useful with no ability for VHF. You aren't going to get much range out of UHF with a 5mW output limit.


If you look at other sources, you'll see that the 9363 can be coaxed into being a (usually up to spec) 9374: https://www.rtl-sdr.com/adalm-pluto-sdr-hack-tune-70-mhz-to-...


$250 from Analog directly, $200 from Amazon right now. Where can I find it for $150?


Most of the major component distributors: https://www.findchips.com/search/Adalm-pluto


I've only used the RTL-SDR ones, so I can't offer any first-hand comparison. But for receive-only operations, the RTL-SDR dongles are cheap, readily available, compatible with many or most SDR related software packages, and basically "just work". YMMV, but one of those is probably where I'd suggest starting.

I've heard good things about the HackRF devices as well. The nice thing about those is that they're programmable transmitters as well, not just receivers. But they're a bit pricier as well, which is the downside. If I buy another SDR anytime soon, that will probably be the direction I go.


If you just want to _look_ at signals, a RTL-SDR kit with a basic dipole antenna (rabbit ears) and the USB dongle is like $35USD and readily available.

I would 100% recommend starting there. If you actually find yourself using it and/or enjoying it, then look at spending some more money.

If you do enjoy playing with SDR stuff, could also recommend getting an amateur radio license. Basic license is ridiculously easy to get and gives you access to legally transmit on a whole bunch of different frequency ranges and put you in contact with some other RF geeks.


I'm a big fan of my LimeSDR Minis: https://www.crowdsupply.com/lime-micro/limesdr-mini/

The SDR comparison table on that page above is useful too.


You can sometimes find a good deal on a HackRF as well. It has a freq range of 1mhz to 6ghz (rx/tx). The popular RTL-SDR only has 0.5mhz to 1.7ghz freq range (rx), but costs only about 1/4.


If you're okay with (1) not supporting the guy that designed it in the first place and (2) taking a risk on a Chinese knock-off (reviews generally seem good), you can often find the Hackrf One on the usual suspect sites straight from China relatively cheaply.

There's a listing right now for just the hackrf one board at $125USD (versus a legit one at $350USD).

Hate to say it but I ordered one from there (with a metal case, high stability clock, assortment of antennas and other useful stuff) for $160.

I've had my eye on one for years, but it's just never going to be something I'm willing to drop $350+ on as part of the hobby.


I don't think you can really call them a knockoff. As far as I know, the HackRF has been designed as an opensource hardware[0], so I'd guess the designer intended it to be built by different entities.

[0] https://github.com/mossmann/hackrf/tree/master/hardware

  "This is a free hardware design; you can redistribute it and/or modify it under the terms of the GNU General Public License[..]"


The SDR you actually own is the best one - the cheap Chinese ones are rubbish, don't get me wrong, but you can still do a hell of a lot with a little RTL-SDR

Analog Devices have some good offerings well up the price scale but I forgot if they start at DC


Depends on what you wanna do with it, but a Red Pitaya is pretty good. Wish I had that instead of the hackRF


https://www.redpitaya.com/f135/spectrum-analyzer

> Frequency span is from DC up to 62.5MHz

This seems rather limiting.


That covers all of the amateur HF bands, so you'll have lots of signals of known types that you can try to decode.


wow what the hell? I did not expect it to only go to 62.5MHz. My bad. I thought it covered all the way up to UHF.


Get an upconverter


very bad idea for many reasons


Wow this has come a decent way since I first tried it back in early 2017. I used to use the rtlsdr a lot as a hobby back then, and it inspired me to get a ham license after hearing the local chatter. I'll have to dig it back out and scan the airwaves again and see if I can decode anything interesting. Thanks for sharing this.


Is there anything like this, but for serial (ie. UART)?



That looks great, although I thought about something more high level, a software which works with USB-UART converters. Like minicom or HyperTerminal, with built in protocol debugging capabilities.



Also check out the Pulseview GUI front-end for sigrock, it is excellent.


Super valuable. I remember specing out Ettus radios for a security research project and they were in the thousands of dollars, whereas today it's a $20 dongle with gnuradio.


Well, the Ettus radios are still probably superior if you’re working with higher frequencies, but there are still cheaper alternatives.

Last time I was at the Ettus site I was dumbfounded more with how much they’re trying to sell software for.


My comment was a fragment of something I thought I didn't post, it's there in error. It was in reference to some RF security work I did almost 10 years ago. Ettus seems super cool, and they were more cost effective than that, and this was just before people popularly discovered that TV tuners could be used as SDRs.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: