Alas, wired lans can be physically secured (more so if they're on a
"high fiber" diet), and to the extent that they are, the data in them
is afforded a resulting level of security.
This is not true of 802.11 with WEP.
i don't think you're right about this, and i've looked into the
history of 802.11 quite a bit. there is some revisionism going on,
and i don't buy it.
the original document describing WEP, by Kerry Lynn, P802.11-94/22,
"IEEE 802.11 Wireless LAN Medium Access Control and Physical Layer
Specifications, The RT Data Confidentiality Algorithm", 7 March1994,
cites IEEE P802.11-94/9, "Radio Equipment and Systems (RES); HIPERLAN
Security Information (input for STAG)", 14 December 1993 as defining
"security comparable to that of a wired LAN" as at least protecting
authorized users of a wireless LAN from casual eavesdropping and data
injection.
WEP is weak enought that it at *most* protects against casual
eavesdropping and not against data injection (if memory serves based
on the Borisov findings), and that's only if you define "casual" as
something like "unwilling to install software and having only the
patience of 2 year old".
Moreover, the original Lynn paper claims the confidentiality provided is
Strong: The security afforded by the algorithm should rely on the
difficulty of discovering the secret key through a brute-force
attack. This in turn is related to the length of the secret key
(usually expressed in bits) and the frequency of changing keys.
However, it may be an easier problem to discover k through
statistical methods if the key sequence remains fixed and
significant quantities of ciphertext are available to the attacker.
RT avoids this by frequently changing the IV and hence k.
As it turns out, he was wrong. It's weak, independent of the security
of the secret key, simply due to the IV reuse.
Although the author seemed aware of known plaintext attacks, he was
unaware of the IV reuse issue, which my cryptographer friends tell me
is a widely-known and fundamental property of stream ciphers, and
was so in 1994.
The last sentence of the paper says what the true agenda was of
the proposal:
"This submission proposes a canditate algorithm that strikes a
balance between the conflicting contstraints of strength, efficiency,
and exportability" [sic]
So my interpretation is:
They wanted something guaranteed exportable, and they didn't get
adequate cryptographic review.
On Wed, Jun 26, 2002 at 12:17:10PM -1000, Paul C. Lawler wrote:
> That is correct. It's design goal was to provide security equivalent to
> a unsecured "wired" line. Nothing more than that. No enhanced or
> additional security exists in the original IEEE specs for WEP.
>
> At 02:56 PM 6/26/02 -0700, [EMAIL PROTECTED] wrote:
> >pardon me, but wep does not work "as designed", unless you are
> >claiming its design goal was to not provide any real security over
> >non-wep.
> >
> >wep is a better example of "the failure of the standards process"
> >than of "works as designed".
> >
> >wep's inexcusable flaw is that 40 bit wep provides the same level of
> >security as 128 bit wep due to the reuse of the rapid reuse of the
> >keystream of a stream cipher.
> >
> >i entirely expect that manufacturers who made claims that 128 bit wep
> >provided enhanced or additional security will be the subject of class
> >actions in the next few years. (at first they were wrong, but at some
> >point, they were just lying to consumers.)
> >
> >
> >On Wed, Jun 26, 2002 at 10:42:24AM -1000, Paul C. Lawler wrote:
> >> I know that at least 4 of them offer additional security (read, more
> >than
> >> WEP) on top of wireless.
> >>
> >> As you probably know, WEP (wired equivalency protocol) was only
> >designed
> >> to make 802.11 "as" secure as a wired connection, which is of course,
> >not
> >> very secure.
> >>
> >> WEP works "as designed," but most wireless providers understand the
> >need
> >> for "real" security in addition to WEP.
> >>
> >> At 12:48 PM 6/26/02 -0700, you wrote:
> >> >* plawler ([EMAIL PROTECTED]) wrote:
> >> >>Boingo, WiFiMetro, Wayport, iPass, T-Mobile (just to name the big
> >ones).
> >> >
> >> >they may do wireless, but those aren't security companies
> >> >
> >> >--
> >> >NYCwireless - http://www.nycwireless.net/
> >> >Un/Subscribe:
> >http://lists.nycwireless.net/mailman/listinfo/nycwireless/
> >> >Archives: http://lists.nycwireless.net/pipermail/nycwireless/
> >>
> >> --
> >> NYCwireless - http://www.nycwireless.net/
> >> Un/Subscribe:
> >http://lists.nycwireless.net/mailman/listinfo/nycwireless/
> >> Archives: http://lists.nycwireless.net/pipermail/nycwireless/
>
> --
> NYCwireless - http://www.nycwireless.net/
> Un/Subscribe: http://lists.nycwireless.net/mailman/listinfo/nycwireless/
> Archives: http://lists.nycwireless.net/pipermail/nycwireless/
--
NYCwireless - http://www.nycwireless.net/
Un/Subscribe: http://lists.nycwireless.net/mailman/listinfo/nycwireless/
Archives: http://lists.nycwireless.net/pipermail/nycwireless/
