From: lsi (stuart@cyberdelix.net)
Date: Tue May 11 2004 - 20:48:07 EDT
[collated replies to a number of repondents below]
> It seems to me that some of these attacks sound great at first, but break
> down when you consider how it would REALLY play out. For one, if you get on
> the train and inventory everyone's clothing...how do you know which shirt
> goes with which pants or shoes?
Easy - signal strength.
> As for credit cards, this is extremely easy to deal with. The cards
> themselves that have been seen so far have a very limited range, measured in
> inches. I can think of a wallet design that would shield the cards a bit,
It's a plan, but this is just asking for the Black Hat to use a
stronger transceiver. It's just building a higher fence; not really
a long-term solution.
> up against everyone like a comically-indiscreet pickpocket. And this all
> assumes that all the credit cards in the wallet don't respond at the same
> time, on the same frequency, thus garbling the results.
This problem will have been overcome in the design of the RFID, as
they are specifically intended for use in counting large batches of
goods.
> I don't think RFID was ever intended to be a feature of security, but rather
> one of convenience.
Unfortunately, the real world dictates that security be a feature of
pretty well everything.
> Tags have to recive the right signal to transmit the data back. If tags
> could be queried by any device wireless networks (900Mhz) would be flooded
> with 900Mhz tags. Wal-Mart is going with the 915Mhz tags so that problem is
> unacceptable. You have to know what to send a tag to get it to respond.
OK, but say the chips are put into cash. Every till on the planet is
gonna know what frequency to use. So, the Black Hat can make a
transceiver for that frequency. Passport RFIDs will also have a
standard frequency - so he can make his transceiver support that
frequency too. In fact, due to the homogenised nature of modern
commerce, there may well only be, in the end, a small number of
frequencies that are used. So Black Hat will end up with the RFID
equivalent of an autorooter. It knows which frequencies are for
what, and it tries them all.
> You run the same risks with using a computer or storing data on any device,
> including pen and paper. A tag id by itself is worthless unless you know
> exactly what data is stored on it.
I imagine that a database will be built which will list individual
numbers, and ranges of numbers, which are known to correspond to
specific items.
But wait - you're saying the tag ID is not the only data on the tag?
> A RFid tag has big limitations too, once you chop off part of the antenna
> it's worthless. The physics of radio waves limits that.
>
> You can't be tracked "everywhere" you go. It's not cost affective at all. A
> tag will transmit around 9 feet.
My understanding is that if you pump more voltage into your
transmission, the tag will transmit further.
This is because the tag simply receives the signal and, using the
power in the signal it received, transmits its ID. If you exceed the
rated spec by, say, 15%, the tag might transmit 15% further than its
rated range. You might be able to snoop when you should not be able
to. ...Or maybe not - that's what the pen-test is for.
> Tags can be shut down (killed) in a second with a reader.
Sweet.
> Instead of fearing new technology, how about working with it to find the
> best ways to use it?
Well, because someone else will be working to find the best way to
*abuse* it, of course. It makes sense to anticipate this activity,
so as not to be burned by it. Again, this is what the pen-test is
for.
If it was my system vulnerable to attack, I would be scared, yes.
I don't make a habit of "working with" defective products - I try and
replace them!
> Tech. improves every day, either we can work with it, or fight it and go
> back to the stone age. I would rather work with it so I can have the
> challenge of security then not have advancements at all.
Advancement for advancement's sake alone is Bad.
When the "challenge of security" can cost people money, waste their
time, violate their privacy, or otherwise, cause them misery, the
technology should not be used. It's a liability, as the lawsuits
that will follow will surely prove.
Deploying a half-baked solution is just not a good idea.
Stuart
--- Stuart Udall stuart at@cyberdelix.dot net - http://www.cyberdelix.net/ --- * Origin: lsi: revolution through evolution (192.168.0.2) ------------------------------------------------------------------------------ Ethical Hacking at the InfoSec Institute. Mention this ad and get $545 off any course! All of our class sizes are guaranteed to be 10 students or less to facilitate one-on-one interaction with one of our expert instructors. Attend a course taught by an expert instructor with years of in-the-field pen testing experience in our state of the art hacking lab. Master the skills of an Ethical Hacker to better assess the security of your organization. Visit us at: http://www.infosecinstitute.com/courses/ethical_hacking_training.html -------------------------------------------------------------------------------
This archive was generated by hypermail 2.1.7 : Sat Apr 12 2008 - 10:53:54 EDT