search

martinmarinov / TempestSDR

Remote video eavesdropping using a software-defined radio platform
GNU General Public License v3.0
1.27k stars 241 forks source link

Comment Algorithms #14

Closed xloem closed 5 years ago

xloem commented 6 years ago

TempestSDR appears to include some sweet algorithms for analyzing signals to provide for van eck phreaking. However, the details of this code are completely uncommented, making it labor intensive to learn from or expand upon it. Please comment the purpose and method of all functions.

martinmarinov commented 6 years ago

The code is not something I'm particularly proud of :D It was done so it is practical rather than beautiful. Never had the time to properly clean it up.

The only parts that are interesting are:

Those are pretty much the main DSP algorithms used and needed to get this to work. They are a bit spaghettificated in the code I'm afraid.

martinmarinov commented 6 years ago

Important part is also detection of dropped usb frames which is essential for high sampling rates

mgkuhn commented 6 years ago

Martin's Master thesis does explain parts of the the underlying theory and the algorithms used in his implementation in Chapter 3. I am in the process of writing up a more detailed tutorial on algorithms for visualizing compromising video emanations from SDR IQ data (including some newer algorithms not yet found in TempestSDR). I'll add a note here when that paper is ready.

reyesjl commented 5 years ago

Hey which files do you need the comments on?

martinmarinov commented 5 years ago

@madman-bob has created a pull request with some comments that is now merged into the main repo.

Thanks to @madman-bob for his effort and time!

xloem commented 5 years ago

@madman-bob, thanks for addressing as much of this as you did. I posted the bounty to help the baseline of information security a little.

madman-bob commented 5 years ago

I'm glad to have been of assistance. I was certainly intrigued when I came across the bounty - I was unaware of this method of attack, and @martinmarinov's dissertation was fascinating.

Do let me know if you know of any similar bounties.

mgkuhn commented 5 years ago

https://www.bountysource.com/teams/tempestsdr

xloem commented 3 years ago

Martin's Master thesis does explain parts of the the underlying theory and the algorithms used in his implementation in Chapter 3. I am in the process of writing up a more detailed tutorial on algorithms for visualizing compromising video emanations from SDR IQ data (including some newer algorithms not yet found in TempestSDR). I'll add a note here when that paper is ready.

@mgkuhn thank you so much for your maintenance work on this public repository. Did you ever finish the tutorial and paper you mentioned here in 2017?

xloem commented 3 years ago

Hey, I'm so sorry, I am actually on the autism spectrum. I'm not sure what to make of your :eyes: reply: I'm not sure what this means.

Once, in the USA, I contacted a shielded room manufacturer asking if I could purchase plans from them for spaces that would protect against undesired communication including attacks like this. I was hoping to find a way to provide ways for communities to design and build community shielded rooms at makerspaces.

I was informed that I needed paperwork regarding access to classified information to be provided such plans. This surprised me: I was not expecting it.

If you guys might not end up sharing your work and research with the wider community, do you know if it is safe for people to pursue research like this on their own? Or are there laws or organisations preventing it?

mgkuhn commented 3 years ago

@xloem I'm not sure either what these GitHub eyes stand for, but “embarrassed look remembering that promised tutorial still hasn't appeared” would be one legitimate interpretation.

mgkuhn commented 3 years ago

@xloem I don't think shielded rooms are a classified technology. There are plenty of companies selling components, or offer building and/or testing them as a service. It's quite an industry, e.g. every hospital MRI machine stands in an EM-shielded room, for example. Just one example supplier: https://hollandshielding.com/Faraday-cages-standard-custom-made-EMI-shielded-test-boxes

xloem commented 3 years ago

Well, right. They were happy to sell me a SCIF but needed impossible paperwork for plans to one. That's why I ask if it is generally legal and safe for non-academic non-government folk to work on research like this.

Thanks for your replies. Anything you can do to share more of what you guys have is wonderful. Soon we might have our EM spectrum run by AI.