Enemy at the Door

[Steve Schear]

At 12:31 AM 11/8/2001 -0500, Dave Emery wrote:
>On Wed, Nov 07, 2001 at 11:01:10PM -0600, Jim Choate wrote:
> > On Wed, 7 Nov 2001, Harmon Seaver wrote:
> >
> > > Duh!  Read it again. "802.11'd to DSL to a very remote web site?"  That
> > > DSL line could be clear across town.
> >
> > Not with 802.11 anything will it be 'clear across town'. A few hundred
> > yards w/ 802.11b and maybe a mile with 802.11a. Now if you're talking
> > directional then it wouldn't take a rocket scientist to ask "Where does
> > that nifty parabolic point? Why, at that other nifty parabolic. Well,
> > gentlemen, how's 'bout we take 'em both down?"
> >
> > If you want distance you'll either have to add an illegal final or else
> > use packet or some other mechanism (probably illegaly as well).
> >
>
>         I have read reports of people running WiFi links of up to 20
>miles.   Given a clear path (clear line of sight) and relatively modest
>sized directional antennas (not huge suspicious looking dishes) which
>can be concealed under rf transparent radomes (hidden in an attic for
>example with appropriate (fiberglass) roofing or siding, or behind a
>glass picture window with curtains drawn) getting 10 to 20 mile ranges
>is pretty easy with gain antennas on either end... not rocket science
>either... and quite hard to spot visually (though of course a spectrum
>analyzer with good preamps and antennas will find and locate any hidden
>802.11 link in no time flat - one cannot radiate rf from a fixed
>location and not be easily found using common TSCM tools).

While I agree that spotting a low process gain, large, signal is 
straightforward with simple TSCM, locating a low power, high process gain 
signal (especially one which uses channel estimation techniques prior to 
transmission, for example "Low probability of intercept communication 
system," US Pat. No. 5,029,184) can thwart most TSCM tools.

steve