# Is TripleDES 168bit vulnerable to Differential Cryptanalysis?

I was reading this paper. In the table-1 of the paper it says TripleDES (showing both 128bit and 168bit keysize) is vulnerable to Differential Cryptanalysis. It also says that AES (showing all three keysize) is strong against Differential Cryptanalysis. If AES-128bit is secure against Differential Cryptanalysis then why not TripleDES-168bit as well?

So, is it the paper that the authors haven't made it clear or is it true that TripleDES 168bit keysize is vulnerable to Differential Cryptanalysis attack?

-
The question is valid, but if that's a scientific paper then I'm little red riding hood. – Maarten Bodewes Jul 18 '14 at 11:54
One should not trust a table/paper where the time to enumerate all 56-bit DES key is given as 400 days at a rate intended to be realistic (the EFF cracker did that in few days in 1998).. and where for 112-bit 2keys-3DES, all other things being equal, the time is only twice that!! $\;$ – fgrieu Jul 18 '14 at 13:07
@owlstead: Agreed. Not only is it full of both factual and grammatical errors, but most of the text, or at least the more fluent parts of it, appears to be plagiarized. If this were an undergrad course assignment, I'd give it an F; since the authors appear to be grad students, and since it appears to have been published in something pretending to be a scientific journal... – Ilmari Karonen Jul 18 '14 at 15:31
@e-sushi, may be its because they are written in uppercase instead of lowercase. Using the uppercase address I also found 404 error. – Giliweed Jul 18 '14 at 16:35
@Giliweed Darn, I would’ve expected Google to rewrite such URLs accordingly. My mistake… thanks for the heads-up on that! Much appreciated. (Removed my related comment to avoid confusion.) – e-sushi Jul 18 '14 at 16:37

## 1 Answer

This claim is bogus. DES itself has a 13-round differential with probability around $2^{-47}$, so TripleDES with its 48 rounds is resistant to any sort of differential attack.

The paper authors are not really confident in the subject.

-
so, this clears that TripleDES is not vulnerable to Differential Cryptanalysis. ---Thanks. – Giliweed Jul 18 '14 at 12:38
Source: Wikipedia: "In 1994, a member of the original IBM DES team, Don Coppersmith, published a paper stating that differential cryptanalysis was known to IBM as early as 1974, and that defending against differential cryptanalysis had been a design goal.[2]" – Maarten Bodewes Jul 18 '14 at 13:40
They seem quite confident, the problem is they're not so competent. ;) – Digital Chris Jul 18 '14 at 14:37