-2

I am writing a paper on RSA, and I am calculating some encryption running times.

I have been told that the encryption running time is $O(n^2)$ where $n$ is the modulus. I have also been told that there is a constant $C$ such that $O(Cn^2)$ where $C$ relates to the machine used.

Does anyone have any idea where to find this $C$ for a certain computer?

e-sushi
  • 17,891
  • 12
  • 83
  • 229
Mathematica
  • 13
  • 2
  • Ok, but will the little-o notation give me the running time for RSA encryption? I am trying to find an upper bound for the public key exponent, so I decided on a max. time - the other factor is how quickly my computer can handle the computations (right?). – Mathematica Nov 10 '15 at 13:10
  • 1
    You seem to confuse public exponent ($e$) and public modulus ($N$, with $n$ the number of bits in $N$); and have trouble putting Big-O notation to practical use. This answer might help for both. The software running on your computer matters more than your computer does. – fgrieu Nov 10 '15 at 13:27
  • I am not confusing them. But if I set an upper bound for N, I am effectively setting an upper bound for e because e cannot be greater that the totient of N. Thank you for your answer. – Mathematica Nov 10 '15 at 13:57
  • 4
    @Mathematica Just benchmark different key sizes. – CodesInChaos Nov 10 '15 at 14:29
  • 2
    By definition $O(n^2) = O(Cn^2)$ for any constant $C$ so the notation is at least not quite right. – otus Nov 10 '15 at 16:42

1 Answers1

4

You don't seem to understand what complexity means. $O(Cn^2) = O(n^2)$, so “finding $C$” is not a valid question – any positive $C$ will do.

I recommend looking up some basic complexity theory introduction. There are many introductory books on the subject which can help you understand what this means.

e-sushi
  • 17,891
  • 12
  • 83
  • 229
dionyziz
  • 573
  • 3
  • 11