[Cryptography] Defeating timing attacks

[Tom Mitchell]

On Mon, Jul 17, 2017 at 10:46 AM, Paul Renault
<thaumatechnician at gmail.com> wrote:
> On 2017-07-16 11:55, Arnold Reinhold wrote:
>
> A careless repair technician can defeat a Faraday cage
......
> A Faraday cage made out of metal mosquito screens would be plenty fine
> for almost all computers.  I've seen some enclosing high-security
> government rooms or parts of rooms.  In their case, the screen's
> material was copper.  Pure, fingerprint-covered copper.

Faraday cages work both ways... a nearby geo-science center has a room to
measure the latent magnetism in rock samples.   It is unclassified
and is a wood frame room within a room.  Copper mosquito screen dominates.
I will ask more but I believe there are no nails  that would conduct
magnetic interference through and into the copper cage.  Power in is
also filtered and choked.
Anyone that has had to participate in  a case design to pass Part 15
of the FCC testing plan cold build and test such a room.
Signals move through free air as well as conduction in power cables
and MuMETAL® can be used to shield from magnetic signaling (in CRT displays).

As hard as some cases are to get correct the concepts are easy.
Hand made probes for modern  o-scopes make locating issues easy. Today
logic boards have ground planes with top to bottom stitching  to both
minimize emission at the source as well as protect the logic from
interference.  RF likes surfaces so cables with a multitude of fine wire
conduct RF to ground better than solid wire.    Even explosive factories
give design clues.  One facility I was in had water powered lights.  A
sealed generator
was powered by a flow of water.  As a model,  water flow can be shifted to the
generator or to a drain with no change in flow so the line pressure and
flow would not change.   NRZ data encoding can be filtered to a constant
current flow (no data escapes).

Common literature on stealth tech imply reflective as well as absorbent
materials have value.

Other analysis adds 'whitening' as discussed here to the mix.








-- 
  T o m    M i t c h e l l