I wanted to do research on this topic for my undergrad, but alas no
luck in getting to do it.
I'm no expert on this so take this with a salt shaker :)
For Bluetooth devices:
If I recal the encryption is 128 bit, and I imagine it is the same for others.
In my opinion the only thing wireless keyboards and mice should be
used for is a gaming system. I am a firm believer in "if I can see
it, then I can read it", and with these devices I can atleast sniff
the air and see your keystrokes. I would say go with corded keyboards
and mice for security.
And yes there have been papers written on doing many different types
of attacks on these devices.
-Kinnell
On Tue, 22 Mar 2005 16:13:16 -0700, Badger, Jared
<Jared.Badger_at_acs-inc.com> wrote:
> Hello All,
>
> I was wondering if anybody out there has researched security of wireless
> keyboards. Although I'm sure many people have very interesting opinions,
> what I need is solid technical information.
>
> SCENARIO:
>
> A sneaky eavesdropper, Eve, would like to find out some gossip on her
> neighbor, Alice. Eve knows that Alice uses Yahoo for her email, and would
> just love to be able to log in to Alice's Yahoo mail account. But she
> doesn't know Alice's password. But luck, it seems, may be in Eve's corner,
> as Alice has recently installed a snazzy new wireless keyboard/mouse combo
> on her home computer. Eve, using her formidable knowledge of radio and
> electronics, sets up an antenna to pick up keyboard transmissions from
> Alice's house. After some persistence, Eve's laptop records a username,
> alice_at_yahoo.com and a password, "opensesame". Eve, ecstatic, logs on to
> Yahoo with the captured password and reads Alice's email, discovering lots
> of juicy secrets.
>
> Although this Alice/Eve scenario is fictional, it seems plausible. For a
> true story, see:
>
> http://www.pcworld.com/howto/article/0,aid,108712,00.asp
>
> My job involves reviewing computer security at a bank, and I was very
> surprised to see that nearly all of the computers at one of my branches are
> using these wireless mouse/keyboard combos. It seems like this could be a
> potentially serious security risk, so I would like to do some research on
> this topic. If these manufacturers have incorporated strong security
> measures, then I would like to know what they are. Or if not, then it would
> be better to know than not to know so as to take appropriate precautions.
> (see the above PCWorld story. Note that only 4000 combinations are used,
> trivial for a computer to crack) Of particular interest to me are:
>
> 1. How possible/easy/difficult is it to eavesdrop and capture keystrokes
> from a wireless keyboard using passive means only? What equipment/expertise
> does this require? (I am thinking it would probably take at least a spectrum
> analyzer, receiver, a laptop, and some custom software) What about taking
> the keyboard apart and reverse engineering it?
>
> 2. How easy/difficult would it be to take control of a computer without
> having physical access to the keyboard at the console? What
> equipment/expertise would this require? (Probably at least the same as
> above, plus a transmitter)
>
> One example of a wireless keyboard/mouse combo is displayed here:
>
> http://www.microsoft.com/hardware/mouseandkeyboard/productdetails.aspx?pid=0
> 14
>
> By entering the following FCC ID's into the FCC website, you can get quite a
> bit of interesting information.
>
> FCC ID's: C3KKB9 (keyboard), C3K1008 (mouse)
>
> https://gullfoss2.fcc.gov/prod/oet/cf/eas/reports/GenericSearch.cfm
>
> There are many docs, including photos and lab tests, on the associated
> pages. For example, FCC docs show that this particular keyboard transmits on
> a frequency of 27.095 - 27.195 MHz. From the internal photos, it doesn't
> seem there are enough electronics to perform advanced encryption.
>
> Certainly somebody knows how to do this. Has anybody tried? Been successful?
> Failed? Any information on common manufacturers (Logitech, Microsoft,
> Kensington, etc.), commonly used encryption/decryption, frequencies,
> encoding, signal power and range, etc. would be most appreciated.
>
> Thanks in advance,
>
> Jared Badger, CISSP
>
Received on Mar 23 2005
Intro
