If you're NOT that Dave Emery, well, please purge the brain cells that this little message has eaten up!
TD
From: Dave Emery <[EMAIL PROTECTED]>
To: Richard Crisp <[EMAIL PROTECTED]>
CC: [EMAIL PROTECTED]
Subject: Re: Fw: RE:Confiscation of Anti-War Video
Date: Mon, 28 Oct 2002 18:29:13 -0500
On Mon, Oct 28, 2002 at 02:56:29PM -0800, Richard Crisp wrote:
> hmm, it seems to me that video riding in a stegano fashion on a cell phone
> call would exceed the bandwidth capability of the channel. It's one thing to
> send a single image steganographically on a cell call, but it is another to
> send a live video feed.
> rdc
There are many issues here... in general should this kind of
thing happen often, there will be either CALEA access at the switch to
lists of all traffic from a cellsite or geographical area (remember E911
location capability) or local off air gear deployed to monitor traffic.
Clearly the spooks have developed some modern off-air digital
traffic monitoring stuff - they need it it places without CALEA - and
there is little doubt that it only costs money to hire some programmers
to make it work with US carrier unique variations of protocols.
And needless to say, most of this stuff consists of tuners
coupled to DSPs coupled to high power high end laptops so most all
of the heavy lifting is in code for the DSPs and laptops that sort
through the bits and signals - and upgrading for the domestic
environment is just loading some new code...
And as for using 802.11 stuff, sure it will work the first time
or two it is tried, but soon enough portable 802.11 stuff will appear on
the other side, including both capture and jamming equipment. And most
of the code for capture and analysis can be had from the public domain
from hacker sources right now - not much expense in adapting it.
And locating 802.11b stuff by using appropriate direction
finding sniffer hardware with DF antennas is a pretty obvious thing to
do - presumably finding the points talking to each other should be easy,
making finding the remote relay site and pouncing on it pretty easy.
Also, you are very right about the bandwidth issues, and another
point not made is that the network is a sealed box with analog voice
in and analog voice out rather than a general purpose TCP/IP bit
pipe - one cannot just pervert a voice call into carrying hidden data
on digital cellphones, at least at much bandwidth as the voice is
vocoded into 4800 baud or thereabouts (modern phones use variable
rate stuff) and it would take some truly hairy DSP code to synthesize
a bizarre stream of voice like audio that allowed access to even
a small fraction of that 4800 baud. I suppose it is theoretically
possible, but much harder than the V90 stuff and heavily dependent
on very precise knowledge of the vocoders and error correction used.
One might get 1200 baud or so, but even that would take a lot
of tricks at both ends and sure would not be stenography in any
normal sense in that it would sound like something from mars and nothing
like normal speech in any known language. And probably in order to
get reasonable band with the talk spurt statistics would have to be
very wrong which would make it stick out like a sore thumb anyway.
And what use is 1200 baud when dealing with video ?
And of course data as opposed to voice calls would be targeted
by the CALEA stuff from the git-go since they would actually be the
logical pipe to ex filtrate the video and one with reasonable
bandwidth...
_________________________________________________________________
Get faster connections -- switch to MSN Internet Access! http://resourcecenter.msn.com/access/plans/default.asp
