Satellite eavesdropping of 802.11b traffic
Dirk-Willem van Gulik
dirkx at webweaving.org
Sun May 30 16:37:24 EDT 2004
On May 27, 2004, at 12:35 PM, John Kelsey wrote:
> Does anyone know whether the low-power nature of wireless LANs
> protects them from eavesdropping by satellite? Is there some simple
> reference that would easily let me figure out whether transmitters at
> a given power are in danger of eavesdropping by satellite?
>
If you assume a perfect vacuum (and note that the athmosphere is fairly
opaque at 2.4 Ghz) and perfect antenna's etc - then the specific
detectivity needed in space suggests a not unresonably sized (m2's) and
cold antenna (below 180k) by very resonably NEP which is commercially
available. Given the noise from the earth background (assuming a black
body radiator) at 2.4, the Sun and the likelyhood that that largish
antenna catches a fair chunk of exactly that then you are at the edge
of what would be realistic. However with some clever tricks and
processing, like a phase array, you certainly should be able to at
least detect that short (1-2mseconds) 100Khz wide 2.4Ghz transmisison
at 0.1 watt is happening - assuming you know where to look. Listening
in over a country-sized swath over a prologned periods of time is an
entirely different story. Given that you then need to be at least 3-4
order's of magnitude better - and that you only get at best square root
when increase the easy things like detector size etc, at best - my
guess would be that some flying or earthbound is a heck of a lot
cheaper and more realistic.
There are some good papers on Lidar and Radar detections of clouds in
the 3Ghz range at 12km which should give you more of an idea of the
spatial resolution you could accomplish. When looking at these - bear
in mind that the 2-3kWatt used is reflected by the ice particles - so
what gets back is 30-40dBZ less - and that you can use a phased locked
loop amplifier easily.
Dw
---------------------------------------------------------------------
The Cryptography Mailing List
Unsubscribe by sending "unsubscribe cryptography" to majordomo at metzdowd.com
More information about the cryptography
mailing list