Just something I considered recently -- TEMPEST attacks are pretty well known; basically you intercept EM leaks from a computer to discover compromising information (i.e. encryption keys, what's being displayed on the screen, what's being typed, etc).

Could a three letter agency potentially create a "TEMPEST Dragnet" of sorts, similar to the NSA's Xkeyscore? Would it be possible to use, say, cell tower antennas to listen to EM emissions and record what's being displayed on thousands of peoples' computer screens, or record everyone's PGP or SSL keys, and create a Xkeyscore-like database?

Theoretically, that would provide them with near total access to peoples' computers -- whoever did it would be able to know exactly what they were doing, both online and offline, and decrypt anything they sent over the Internet. But would it actually be possible?

I'm not an expert on this... can anyone determine the feasibility of such a thing? I suppose the largest concern would be interference -- it would be difficult to tell which emissions belonged to each individual. Would it be possible to filter out or segregate emissions?

  • 1
    inverse square law applies en.m.wikipedia.org/wiki/Inverse-square_law – Neil Smithline Sep 13 '15 at 15:25
  • The main problem is cost. Side emissions are low-power and not concentrated in one particular band - there's a wide variety of CPU, bus etc. clock frequencies, and high-gain, low-noise, very wide band receivers cost a lot, and furthermore must be set up very close to targets. – Deer Hunter Sep 13 '15 at 17:11

Given current technology, no, I don't believe this is at all practical. The problem is there is a lot of EM out there and you would need a way of being able to make sense of all that noise. The techniques you refer to normally require close proximity to a specific target and the technique normally needs to be tuned to that particular target because different types/brands/models of electronic equipment have different EM signatures.

Think of a busy room full of people talking. You can't easily hear and understand all those conversations - they are just noise. However, if you focus on a specific conversation, then you may well be able to pick up understandable content. However, you will be limited to those closest to you - it isn't easy to filter out all the closest speakers and just focus on someone across the room. If you wanted to pick up all the conversations, you would need to have nearly as many people as conversations and theyw ould need to be spread throught the room. So, you would need lots of resources and then, for this information to be of any use, you would need to post-process the data you collect.

Of course, this doens't rule out someone finding a way in the future where they can isolate the signals and increase the distance you are able to pick up the signals - so it may be possible in the future, but I don't believe it is practical now.

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.