Extracting a 3DES key from an IBM 4758

Part 7: Who are we ?

We are two research students in our second year of a PhD course in the Security Research Group of the Cambridge University Computer Laboratory .

	Mike Bond's PhD thesis will be on cryptoprocessor APIs. He invented this attack on the IBM 4758. It is related to, but not the same as a number of attacks that were described in his CHES 2001 paper. Readers may also be interested in a more general paper co-written with Dr Ross Anderson in the October 2001 edition of IEEE Computer magazine.
	Richard Clayton's PhD will be on an unrelated topic, which will draw upon his many years in the ISP industry. He worked on this particular project just for fun! His contribution was the FPGA design that was used in the attack. Using hardware is very much faster then using programs on general purpose computers. It means that the attack can be mounted for $995 in two days, rather than requiring a $3000 high speed PC for two months. Richard's web page that collects together a whole heap of information on brute force attacks on crypto systems -- and on DES in particular can be found here.

At an earlier stage of this project, when we were considering this attack, but many details were still to be settled, we gave a talk as part of the Security Group Seminar Series. The slides from this talk can be found here.

We'd like to acknowledge the generosity of Altera in providing the FPGA board used in this project for free. We'd also like to thank Ross Anderson and Simon Moore for their helpful comments and encouragement throughout.

Finally, if anyone has access to an IBM 4758 in a real world application (the more valuable the data it's transferring the better) we'd be delighted to have the opportunity to run our attack "for real" :-)

Next part: Do It Yourself !
Previous part: Some real results