Computational cryptography › Multivariate
SIGNATUREPrimitiveHistorically broken

SFLASH

In plain words

A fast multivariate signature for smart cards, recommended by Europe's NESSIE project in 2003 and broken in 2007.

The idea: SFLASH hid part of C*'s output (the "minus" modification) so Patarin's equations would stop working. The 2007 attack found that the hidden map still had a symmetry that could be used to rebuild the missing outputs, after which Patarin's old attack worked again.

Why it matters: A reminder that a standards body's approval isn't proof of security.

The details

On the map: ✗ recommended by NESSIE (2003)

A fast, short multivariate signature built on C* with the minus modification. The European NESSIE project recommended it in 2003 for low-cost smart cards[1]. The 2007 attack was practical on an ordinary computer[2].

Like SIKE and Rainbow, it shows that a standards body's approval is not proof of security.

Rests on

If any of these breaks, this breaks too.

If it falls

If this breaks, these break with it.

Nothing on the map depends on it: it is an end point, something people use directly.

References

  1. NESSIE consortium. Portfolio of recommended cryptographic primitives. NESSIE project, 2003. cosic.esat.kuleuven.be/nessie/deliverables/decision-final.pdf
  2. Vivien Dubois, Pierre-Alain Fouque, Adi Shamir and Jacques Stern. Practical Cryptanalysis of SFLASH. Advances in Cryptology - CRYPTO 2007. doi:10.1007/978-3-540-74143-5_1