A new attack on RSA and CRT-RSA

  • Authors:
  • Abderrahmane Nitaj

  • Affiliations:
  • Laboratoire de Mathématiques Nicolas Oresme, Université de Caen, France

  • Venue:
  • AFRICACRYPT'12 Proceedings of the 5th international conference on Cryptology in Africa
  • Year:
  • 2012

Quantified Score

Hi-index 0.00

Visualization

Abstract

In RSA, the public modulus N=pq is the product of two primes of the same bit-size, the public exponent e and the private exponent d satisfy $ed\equiv 1 \pmod{(p - 1)(q - 1)}$. In many applications of RSA, d is chosen to be small. This was cryptanalyzed by Wiener in 1990 who showed that RSA is insecure if dN0.25. As an alternative, Quisquater and Couvreur proposed the CRT-RSA scheme in the decryption phase, where $d_p = d \pmod{(p - 1)}$ and $d_q = d \pmod{(q - 1)}$ are chosen significantly smaller than p and q. In 2006, Bleichenbacher and May presented an attack on CRT-RSA when the CRT-exponents dp and dq are both suitably small. In this paper, we show that RSA is insecure if the public exponent e satisfies an equation $ex+y\equiv 0\pmod p$ with $|x||y|dp say, satisfies $d_pdp and dq are required to be suitably small.