Re: Null integrity protection and encryption algorithms



On May 7, 4:21 am, mike clark <undefinedsp...@xxxxxxxxx> wrote:
But, I wonder why should we need to go in for an algorithm to make
'b' appear as 'b'  or apply an algorithm to get a 32-bit MAC of all
zeros ?
Isn't it enough to send it directly without applying any NULL
integrity protection
or NULL encryption algorithm ?

Any guesses? What if you have a predefined protocol?

Okay.

But anyhow,
One of the reason for this query is because this algorithm inturn
consumes some processor instruction cycles and there is an
impact on performance/energy consumption. So, isn't it enough
to send without applying either the NULL integrity protection or
NULL encryption algorithm if it were to behave such that
NULL(b) = I(b) = b ?

I wonder Why should we need to go in for an algorithm to make
'b' appear as 'b' or to generate a 32-bit MAC of all zeros .
Why can't we just make a copy of 'b' to the output of the encryption
directly instead of doing the NULL encryption algorithm to get 'b'
as encrypted output ?
Also, similarly, why can't we fill the 32-bit MAC with all zeros
instead of
applying the NULL algorithm to get a 32-bit MAC of all zeros ?

By this method, we can improve the performance becuase using
an algorithm will consume a cycle . But, i wonder why they came
up with a NULL algorithm to make 'b' to appear as 'b' or to fill zeros
in 32-bit MAC ?

Thx in advans
.