Talk:CMAC

Article name available?

"It has been suggested that this article or section be merged with One-key MAC."

Does that mean the entry "CMAC" would be free to use for the Cerebellar Model Articulation Controller (a type of neural network classifier) ? Dcornforth 09:59, 28 November 2006

Yes, once the merge has been done it should be okay to use the name for the neural network thing, especially if we ad a disambig link at the top of the article to the One-key MAC article and fix all incoming crypto related links so they go to the One-key MAC article instead. For now I suggest you use the title Cerebellar Model Articulation Controller and ad a disambig link at this article (CMAC) to the neural article. Sorry that I don't have time to do the merge. I hope some other crypto editor has the time some day. --David Göthberg 03:56, 5 December 2006 (UTC)

The CMAC neural net concept dates from 1975, the crytographic algorithm apparently only from 2006 or so? Qwertyzoop (talk) 18:09, 29 April 2010 (UTC)

I agree that the topics "One-key MAC" and "Cipher-based Message Authentication Code" (CMAC) are so closely related that they should be merged into a single "One-key MAC" article.

After that is done, we can discuss whether "CMAC" should be a redirect to that merged article, or to the "Cerebellar Model Articulation Controller" article, or to the "CMAC (disambiguation)" article. --DavidCary (talk) 15:02, 20 May 2016 (UTC)

Finite field?

$C=1010_{2}$ in the example does not give rise to a finite field, because $x^{4}+x^{3}+x$ is not irreducible modulo 2. As a consequence the low-order bit of $k_{1}$ and $k_{2}$ will always be 0. Surely this is not intended? There ought to be a remark in the article that $C$ must be chosen with some care to ensure that entropy is not lost in the multiplication step. –Henning Makholm 11:33, 5 May 2007 (UTC)

In fact the choice specified by NIST 800-38B implies that the four-bit $C$ must be $0011_{2}$. I will change the example accordingly. –Henning Makholm 11:49, 5 May 2007 (UTC)

k1 and k2 generation, step 2

"... namely the non-leading coefficients of the lexicographically first irreducible degree-b binary polynomial with the minimal number of ones"

Could someone please rephrase this to make it clearer and easier to understand, giving an example if appropriate? —Preceding unsigned comment added by 192.54.144.229 (talk) 12:14, 12 February 2009 (UTC)

I think the next easier formulation is "a certain b-bit constant". There's quite a bit of abstract algebra going into the precise selection of C, and trying to explain it here would be irrelevant to the article. I have edited the description to clarify the central fact that C does not depend on the key, so in any practical application the appropriate C can be computed once and for all and included in the source code as a constant. This should probably be made clearer yet, but I'm not quite sure how to do it without disrupting the flow of the article.

As for an example, it is already there: When b=4, C becomes 0011₂. –Henning Makholm 00:15, 15 February 2009 (UTC)