GSM - A5 Strength (was Inaccurate study quoting, Re: anti-crypto rhetoric)

[Brian Gladman]

From: Casper Dik <casper@holland.Sun.COM>
To: ukcrypto@maillist.ox.ac.uk <ukcrypto@maillist.ox.ac.uk>
Date: 07 April 1998 15:09
Subject: Re: Inaccurate study quoting, Re: anti-crypto rhetoric (Ellison,

>>About GSM
>>
>>On Tue, Apr 07, 1998 at 12:26:18PM +0100, Phillip Temple wrote:
>>> users in 106 countries, over 1/4 billion by 2003. They are already using
>>> secure crypto, with no government back-doors, designed to withstand
>>> a sustained crypto attack. However...
>>
>>I don't know if you could describe A5 (GSM crypto algo) as "strong".
>>
>>http://chem.leeds.ac.uk/ICAMS/people/jon/a5.html
>>
>>I have been told, and you may want to treat this with a pinch of salt,
>>that some of the US networks using PCN have the encryption turned
>>off for performance reasons.


There are many on this list who know a great deal more than I do about GSM
but I suspect that the A5 algorithm only protects the communications between
the phone and the nearest tx/rx node.  I would guess that data in the
swtiches, and that in transit between them, is not cryptographically
protected.

Although I have no direct knowledge of the strength of A5, my suspicions are
that it is very weak because it has been (and, I think, still is) the policy
of the UK government (and I guess many others) not to allow strong
cryptographic algorithms to go into widespread use.

         Brian Gladman