From: doug@netcom.com (Doug Merritt)
Date: Tue, 12 Oct 93 20:36:35 PDT
To: cypherpunks@toad.com
Subject: Re: Spread-spectrum net (vulnerability of)
In-Reply-To: <mg5n+@andrew.cmu.edu>
Message-ID: <9310130301.AA12952@netcom5.netcom.com>
MIME-Version: 1.0
Content-Type: text/plain


Matthew J Ghio <mg5n+@andrew.cmu.edu> said:
>Depends how you place them. If you put them _on top_ of things, you'd need 
>a helicopter to shoot 'em.

I used to work with the person who set up 80% of the West Coast ham
radio digital packet system repeaters. He had a bit of money to burn,
and he set up a *bunch* of these repeaters on various mountain tops up
and down the West Coast. This was circa '85 by the way.

Each repeater required:
	1) a source of power -- this is easy to forget, but it is
	   nontrivial. Batteries just don't cut it. Rather than explain
	   about batteries, I invite people to ponder power requirements
	   for given broadcast distances and guess battery lifetime.

	   Anyway, he had a PG&E (electric power) hookup and a monthly
	   bill for each of these. Good thing he had all that money...
	   
	   He had trouble finding people who were willing to allow his
	   stations; people who live on mountain-tops tend to be
	   rugged individualists. :-) But he managed.

	2) Line of sight reception and transmission. The higher the
	   frequency used, the more that line of sight is required by
	   the physics of the situation. Even if line of sight isn't
	   strictly required, signal strength drops when you get out
	   of line of sight -- if you depend on radio wave diffraction
	   to get the signal to you and from you, you can count on heavy
	   signal loss in the process.

	3) Protection from the elements. He used rack-mounted 286 cards
	   in an industrial enclosure, but there was *no way* he could
	   leave the boxes exposed to the elements. A roof was required.

	   This is the kind of issue that is easy to neglect, but is
	   absolutely essential for real life conditions: weatherproofing
	   of the assembly. If you don't use as roof and walls, you'd
	   better have a genius mechanical engineer design the enclosure.
	   This is a lot harder than it sounds.

>Someone walking around a city shooting a rifle is likely to attract a
>lot more attention than a secret network would.

Your opponents wouldn't do that...the "enemy" here is simply the FCC,
count on it. They have field agents who triangulate illicit transmitters,
and once they find one, they simply get all the warrants and court orders
needed to deal with it.

>Secondly, the
>transmitter doesn't necessarily have to be exposed, it could be kept
>hidden and only the antenna would need to be exposed.  You'd have to be
>a damn good shot to hit a wire antenna.  Plus the antenna would be easy
>to disguise or hide in many places.

Sure, this is possible. But it doesn't help *that* much. Triangulation
of signal spots any antenna quickly, and they cut that off immediately.
You can get increasingly elaborate about hiding the signal source, and
restoring antennae quickly as they're pinched, but it's sort of a losing
battle unless you assume real time response by the underground lead by
a brilliant EE type.

I am not saying that an underground wireless net is impossible. I *am*
saying that the difficulties are much higher than they may seem at first
blush.

I think doing something like this is possible, and it would have definite
benefits. But anyone moving on this would do well to get in touch with
the existing ham radio crowd who have dealt with the pragmatic issues
involved for the better part of a decade (or more).

>  Hence we have achieved
>our objective - you know how to contact site #3 in netspace - it has a
>cybernetic location relative to other sites, but that tells you nothing
>about it's actual physical location.

Right...in a sense, this is the easy part. Definitely one can hide
physical locations if everything is set up carefully.
	Doug