Tempest attacks against AES: Stealing keys using minimal equipment [pdf]

This was the AES implementation this was tested against:<p><i>The trace below shows our signal for one block of AES-256 encryption running on a SmartFusion2 target. We use OpenSSL&#x27;s implementation of AES on the ARM Cortex-M3 core of the SmartFusion2. There are clear, distinct patterns for each stage of processing. We see I&#x2F;O to and from the Cortex-M3, calculations for the key schedule, and the 14 encryption rounds.</i><p>So it was a software implementation.<p>I wonder if and how effective this attack would be against devices with hardware implementations of AES.

I remember reading about such attacks for the first time in Neal Stephenson&#x27;s book Cryptonomicon under the term &quot;Van Eck Phreaking&quot;. Looks like its gotten a lot easier in recent years!

Are there any modern crypto algorithms that are, by design, immune from an attack such as this? Would not having any key-dependent code paths be sufficient to prevent this attack?<p>If it is possible to be immune by design to power analysis, timing and tempest attacks, is there a list of such algorithms somewhere that I can look it up? My google-fu hasn&#x27;t returned anything useful.

An intelligent noise generator that runs as the second hardware thread on the same CPU using should be able to protect the encryption. If the second noise-generation thread is able to randomly stop the encryption thread and do itself some random crypto, it should be able to fool the eavesdropper which will assume that the signals of the noise thread is produced by the encryption thread.<p>One can also think about modifying the implemenation of OpenSSL and others by inserting a lot of noise in the algorithm itself.<p>One can also ask chip designers to modify the circuitry to produce a lot of noise during AES instructions. Or do the opposite in circuitry: use something comparable to active noise cancellation in headphones.

This is research by my close colleagues; I&#x27;m happy to answer any questions.

Well, in many cases AES keys are used one time, and there&#x27;s also forward secrecy that guards it from decryption even if the key leaked.

There was an attempt to do something similar with ps3 <a href="http:&#x2F;&#x2F;www.eurasia.nu&#x2F;modules.php?name=Forums&amp;file=viewtopic&amp;topic=7266&amp;forum=87" rel="nofollow">http:&#x2F;&#x2F;www.eurasia.nu&#x2F;modules.php?name=Forums&amp;file=viewtopic...</a> , progress stopped though.

Off topic, but I always wondered how defense forces deal with encryption of channel when they collaborate with other forces from different countries. You would somehow be able to add a new participant to the group. Would this require re-issue of keys?

I read it a few times and still don&#x27;t understand how you can get like the 4k of private key data or whatever it is out of a radio signal - and they don&#x27;t even mention keys they&#x27;re talking about the algorithm itself.<p>Totally don&#x27;t get it in the slightest.

alright, portable faraday cages for everyone!

Can someone ELI5 how this works? Would be much appreciated &lt;3

Guys, stop breaking the world! &#x2F;s

I&#x27;m all for the sharing of information and responsible disclosure etc, but when a company that makes stuff that is supposed to be protected from this sort of attack, then shows how if you dont buy their stuff you are at risk from anyone who can follow their plans and has $200, which they likely couldn&#x27;t do yesterday, it doesnt seem to be as consumer friendly as it could be. more protection racket perhaps,